With the proliferation of cloud computing, the exponential growth of data amount requires continuous expansion of storage capacity to meet the storage demand, consequently resulting in higher energy consumption in data centers. However, many of the conventional data placement strategies strive to save energy by optimizing the distribution of data requests, while overlooking the impact of heat recirculation among data nodes. To bridge this gap, we propose a heat-recirculation-aware data placement approach, termed HRADP, designed to optimize data placement in data centers while reducing energy consumption. First, based on the extent of heat recirculation, HRADP places data to the upper limit of the disk allowance for each data node. Second, during energy allocation, HRADP further eliminates disks with high current workloads to prevent localized hotspots. Finally, during request scheduling, HRADP selects dormant data nodes according to the request size, minimizing the startup energy consumption caused by data nodes waking up with small requests. We implement the HRADP approach on a data center simulation platform, CloudSim, and its performance with state-of-the-art, including the Energy-efficient and Thermal-aware Data Placement (ETDP) algorithm, Thermal-aware file assignment technique (TIGER), Storage and rack-sensitive replica placement algorithm (SRS), and Hadoop Distributed File System (HDFS). The experimental results reveal that HRADP revamps the cooling supply temperature, total energy, and data throughput by averages of 0.08%-0.6%, 8.22%-53.03%, and 9.58%-64.59%, respectively.

Programmable switches have revolutionized networking by enabling a new spectrum of applications, such as network telemetry, in-network computation, and machine learning. These applications heavily utilize register memory but their performance is significantly constrained by the scarcity of on-chip resources, such as the 15 MB of SRAM available on a Tofino switch. To effectively accommodate increasingly memorydemanding applications, we aim to pool register resources across multiple switches, creating a larger unified register memory space. This resource pooling approach addresses the limitations of existing single-switch Virtual Register Memory (VRM) solutions, which cannot meet the demands of these applications in distributed environments. To achieve this, we propose dVRM, a distributed VRM deployment framework that enables crossswitch memory sharing and self-dynamic memory allocation on the data plane. dVRM introduces three innovations: (1) Grouped Multi-Switch Registers (GMRs), virtualizing distributed pipeline stages into a unified memory pool; (2) a self-adaptive, bit-width allocation mechanism driven by real-time data-plane feedback; and (3) lightweight heuristics for concurrent application deployment with distributed VRM, formulated as a mixed-integer linear programming (MILP) problem. We have implemented dVRM on both P4 hardware switches (with Intel Tofino ASIC) and BMv2. Experimental results show that dVRM significantly reduces hash unit consumption by up to 26% and achieves an improvement in accuracy (ARE) of up to 57.3% across various workloads.

Communication overhead is a critical challenge in federated learning, particularly in bandwidth-constrained networks. Although many methods have been proposed to reduce communication overhead, most focus solely on compressing individual gradients, overlooking the temporal correlations among them. Prior studies have shown that gradients exhibit spatial correlations, typically reflected in low-rank structures. Through empirical analysis, we further observe a strong temporal correlation between client gradients across adjacent rounds. Based on these observations, we propose GradESTC, a compression technique that exploits both spatial and temporal gradient correlations. GradESTC exploits spatial correlations to decompose each full gradient into a compact set of basis vectors and corresponding combination coefficients. By exploiting temporal correlations, only a small portion of the basis vectorsneed to be dynamically updated in each round. GradESTC significantly reduces communication overhead by transmitting lightweight combination coefficients and a limited number of updated basis vectors instead of the full gradients. Extensive experiments show that, upon reaching a target accuracy level near convergence, GradESTC reduces uplink communication by an average of 39.79% compared to the strongest baseline, while maintaining comparable convergence speed and final accuracy to uncompressed FedAvg. By effectively leveraging spatio-temporal gradient structures, GradESTC offers a practical and scalable solution for communication-efficient federated learning.

Bit-alterable flash-based SSDs, offering page-level erase operation, allows individual flash pages in a block to be erased independently. The page-level erase operation alleviates the overhead of page migration during garbage collection and improves the SSD lifetime. However, when the number of invalid pages within a block exceeds a certain threshold, the latency of page-level garbage collections using page-level erase may exceed that of block-level garbage collections. In bit-alterable flash memory, existing garbage collection strategies dynamically choose between page-level and block-level garbage collections based on their latency. This often fails to fully exploit the advantage of page-level garbage collection in reducing write amplification under low-load conditions.To address this limitation, we propose an adaptive garbage collection strategy called AGCB to dynamically adjust garbage collection operations by the runtime workload of flash channels, thereby enhancing SSD performance and lifetime. Specifically, AGCB classifies flash channels as busy or idle by monitoring the depth of the transaction queue in cache. According to this classification, AGCB selectively applies page-level or block-level garbage collection operations, aiming to minimize the impact of garbage collections with host I/O requests. Meanwhile, we introduce a staged victim block selection scheme to further improve garbage collection efficiency and wear leveling. The experimental results unveil that compared with the existing schemes, AGCB reduces the number of garbage collection operations, average response time, and blocked user requests by an average of 14.6%, 14.7%, and 17.3%, respectively.

In the realm of Mobile Edge Computing (MEC) networks, mobile interactive applications, such as multiplayer online games, federated learning, and interactive multi-view video, are becoming increasingly popular. The embedding of Bidirectional Service Function Chains (BSFCs) for these applications has been studied. However, existing BSFC embedding research only considers static users, and the prevailing Virtual Network Functions (VNFs) placement methods do not account for the impact of individual node resources on the path, thus failing to maximize node resource utilization at the network level. Considering the aforementioned issues, we introduce a framework for BSFC embedding tailored for mobile users, named Path as a Point (PAAP). This framework integrates the resources of the paths and then globally considers the impact of the computing resources of edge nodes on the paths connecting them, maximizing edge node resource utilization across the entire network. We propose a three-phase algorithm within the PAAP framework. In the first phase, we introduce a singleuser algorithm for VNFs deployment during BSFC embedding. In the second phase, this optimization is extended to multiple users by establishing inter-user association rules. In the third phase, candidate application placement positions are evaluated across the entire network, culminating in the determination of optimal placement strategies. Empirical evaluations confirm the effectiveness of the proposed algorithm, demonstrating significant performance improvements over baseline methods.

Mobile Edge Computing (MEC) is an emerging network architecture designed to enhance Quality of Service (QoS) by bringing computational resources and storage closer to end users. In MEC environments, strategic placement of edge servers is crucial to minimize costs and optimize QoS. The Edge Server Placement (ESP) problem has been effectively addressed by using Multi-Armed Bandit (MAB) algorithms, known for their efficiency and adaptability, with the Upper Confidence Bound (UCB) algorithm being particularly prominent due to its stable performance and low dependency on parameters. However, UCB suffers excessive exploration and high uncertainty in reward estimation during its initial phase, leading to slow convergence. To overcome these challenges, we propose a novel Preference-based UCB (P-UCB) algorithm, which integrates the preference function into the UCB framework, drawing inspiration from the Gradient Bandit (GB) method. This modification not only accelerates convergence, but also improves overall efficiency. Furthermore, to address the Base Station Allocation (BSA) issue within the ESP context, we introduce a Weighted Base Station Allocation (WBSA) algorithm, which helps better manage access delay and workload balance. The P-UCB algorithm is evaluated through a comprehensive metric that based on access delay and workload balance, showing significant improvements over existing methods such as Multiple Choice (MC)-UCB, Q- Particle Swarm Optimization (QPSO), and Genetic Algorithm (GA). Experimental results on a real-world dataset show that P-UCB achieves a notable performance increase of at least 12.9%, effectively optimizing access delay and workload balance across various experimental settings, including the number of base stations, edge servers, and other relevant system parameters.

Virtual machine (VM) consolidation strategies are widely deployed in Cloud Data Centers (CDCs) to optimize resource utilization and improve the Quality of Service (QoS). However, the host overload detection algorithms in current VM consolidation strategies are static. That means, once the overload threshold is calculated, it will not change until the next recalculation. The current algorithms are not suitable for the environment of highly dynamic workloads which results in additional energy consumption and potential Service Level Agreement Violations (SLAVs) which will affect the QoS of CDC. In PMPHD, a novel host dynamic threshold adjustment algorithm is proposed. In the proposed algorithm, the PMs are classified into mildly overloaded, normal, and severely overloaded based on the resource utilization. If the PM is predicted to be severely overloaded in the next moment, the threshold of this PM will be proactively reduced. The PM is determined to be overloaded, and some VMs in this PM will be migrated in advance. Thus, this PM will be in normal in the next moment, and the VM performance degradation resulting from SLAV and VM migration overlap in the next moment will be avoided. If the PM is predicted to be mildly overloaded, the threshold will be appropriately increased to transit it to be in normal state in the next moment, and the VM in the PM will not be migrated. Since the PMs’ workloads are dynamic, the PMPHD overload algorithm predicts the resource utilization rate of PM continuously, and adjusts the overload threshold of PM. Compared with other algorithms, PMPHD maintains high efficiency while having lower ESV (a combination metric for balancing energy consumption and SLAV).

With advancements in machine learning (ML) technology and the deployment of large ML-as-a-Service (MLaaS) clouds, accurately understanding request behaviors in an MLaaS cloud platform is paramount for resource scheduling and optimization. This paper sheds light on the correlation of request arrivals in a representative and dynamic MLaaS workload – Alibaba PAI (an ML platform for artificial intelligence). For requests in the PAI workloads at the job, task, instance, and machine levels, our burstiness diagnosis reveals that the request arrival processes at all levels are significantly bursty. Additionally, our Gaussianity test indicates that the bursty activities in PAI consistently appear to be non-Gaussian. Our findings show that there exists a certain degree of correlation between request arrivals at each level over long-term time scales. Moreover, we reveal the self-similar nature of request activities in the various-level wild MLaaS workloads on Alibaba PAI through visual evidence, the auto-correlation structure of the aggregated process of request sequences, and Hurst parameter estimates. Furthermore, we implement a versatile workload synthetic model to synthesize request series based on the inputs measured from the PAI trace. Experimental results demonstrate that our model outperforms typical self-similar workload models, and can improve accuracy by up to 99% compared to them.

Achieving counterfactual fairness is a critical objective in advancing fairness research within machine learning. Studies have shown that machine learning models often inherit biases from their training data, leading to unfair decision-making. Fair data generation methods aim to mitigate these biases, ensuring that predictors trained on such data uphold fairness. However, in the context of counterfactual fairness, existing methods for generating fair data are often limited in their applicability and lead to significant performance losses in downstream predictors. To address these issues, this paper proposes a new algorithm for generating counterfactually fair data, allowing predictors trained on this generated data to adhere to counterfactual fairness. We propose a new metric, Intervention-Driven Correlation (IDC), to evaluate the counterfactual fairness of generative models. IDC assesses fairness by applying random interventions to samples and measuring the statistical correlation between the degree of intervention and the outcome of interest. This metric is applicable to both discrete and continuous sensitive attributes and labels. Furthermore, our studies reveal a critical insight: counterfactually fair data does not always guarantee counterfactually fair predictors when deployed in real-world scenarios. We identify the root causes of this issue and propose a robust solution. To bridge this gap, we propose the IDC-Reduction method, which ensures the fairness of downstream predictors by generating counterfactually fair data. Experimentally, our method outperforms existing approaches and achieves counterfactual fairness regardless of the type of downstream predictors.

In-band network telemetry (INT) enables real-time network monitoring by embedding telemetry data into packets. The advent of programmable switches further enhances the flexibility of INT by enabling dynamic customization of telemetry collection at the hardware level. However, the practical application of INT is hampered by significant challenges arising from data noise (due to packet loss, delay, and measurement inaccuracies) and network dynamics (such as changing INT paths and feature requirements). These issues severely degrade the performance of machine learning models used for analyzing INT data, hindering the accurate capture of spatio-temporal network characteristics. This paper presents Planner, a novel generative graph learning framework designed to address these limitations. Planner enables the collection of network features at various levels of granularity on programmable switches. Crucially, it constructs dynamic graphs representing evolving INT paths and employs a hybrid Graph Neural Network (GNN) and Recurrent Neural Network (RNN) architecture to effectively learn spatial and temporal dependencies. Furthermore, Planner incorporates variational inference to generate robust latent representations, mitigating the detrimental effects of noise and instability in INT data. We have implemented a testbed prototype of Planner using Intel Tofino ASIC switches. Extensive experiments demonstrate the performance superiority and robustness of Planner over the baseline methods, achieving a 23.2% improvement in F1 score.

Sliding window aggregation is a core operation in data stream analysis that extracts summaries from the most recent data stream. An evict or insert of the window can be handled in$O(1)$for in-order data streams. However, real-world data streams are typically disordered due to network delays. To process out-of-order data streams, existing methods primarily use a tree to maintain the sliding windows. Since the complexity of the tree is related to the window size, the performance of these methods will drop sharply or become unavailable when facing a big window. To overcome the limitation of existing methods, this paper presents Finger B-Trees Forest Aggregation (FIFA). This novel forest-based sliding window aggregation scheme optimally handles out-of-order data streams. At its heart, FIFA uses aggregation forest to extend Finger B-Trees. Specifically, FIFA evenly divides a window into several chunks and constructs a separate tree to maintain each chunk. When an out-of-order item arrives, FIFA first locates the corresponding chunk of the item and then uses Finger B-Trees to insert it into the window efficiently. Chunking reduces the complexity of the tree and the coupling of aggregation results by isolating items within windows. Thus, an insert or evict takes amortized$O(c)$in the worst case, where$c$is the size of each chunk. Finally, extensive experiments based on real-world data demonstrate that FIFA achieves an average 2-fold throughput improvement on out-of-order data streams compared with the state-of-the-art (SOTA) aggregation schemes.

The new paradigm of In-network computing (INC) permits service computation to be executed within network paths, rather than solely on dedicated servers. Although the programmable data plane has showcased notable performance advantages for INC application deployments, its effectiveness is constrained by resource limitations, potentially impeding the expressiveness and scalability of these deployments. Conversely, delegating computational tasks to the control plane, supported by general-purpose servers with abundant resources, offers increased flexibility. Nonetheless, this strategy compromises efficiency to a considerable extent, particularly when the system operates under heavy load. To simultaneously exploit the efficiency of data plane and the flexibility of control plane, we proposeCarlo, a cross-plane collaborative optimization framework to support the network-wide deployment of multiple INC applications across both the control and data plane.Carlofirst analyzes resource requirements of various INC applications across different planes. It then establishes mathematical models for resource allocation in cross-plane and automatically generates solutions using proposed algorithms. We have implemented the prototype ofCarloon Intel Tofino ASIC switches and DPDK. Experimental results demonstrate thatCarlocan effectively trade off between computation time and deployment performance while avoiding performance degradation.

The parallel operation technology plays a pivotal role in enhancing performance of 3-D nand flash-based SSDs. High-parallel distribution of consecutive pages places the pages on different parallel units, thereby improving the parallelism and throughput of read requests. However, the high-parallel distribution generates two problems: 1) aggravating data fragmentation and 2) exacerbating the impact of garbage collection (GC) on latency. Moreover, small reads only require a few parallel units, and thus the high-parallel distribution is redundant for the requests. To address this issue, we propose a read-request driven adaptive allocation scheme called RDA to bolster SSD performance by adaptively adjusting the parallel distribution of consecutive pages. The RDA scheme employs the size of historical read requests to gauge the level of parallelism for write requests with varying sizes. Then, RDA allocates the logical pages of writes to distinct parallel units according to the parallelism of the requests. In doing so, RDA effectively mitigates the performance degradation of SSDs caused by redundant parallel distribution, while preserving the parallelism of read requests. We compare RDA with the three state-of-art schemes Amphibian, SOML, and Preemptive GC in terms of GC-blocked read requests, GC counts, and read response time under eight real-world workloads. The experimental results unveil that compared with the existing schemes, RDA revamps the GC-blocked read requests, GC counts, and read response time by averages of 20.6%, 7.8%, and 15.8%, respectively.

Recently, many laws and regulations have granted users the right to be forgotten, i.e., the right to require data controllers to delete user data. Various methods for machine unlearning have been proposed to remove individual data points. However, they do not scale to the scenarios where larger groups of features are to be removed. To address this challenge, we propose MAFRO, an optimal-granularity fuzzy decision rule–based classifier that accelerates unlearning via influence functions. Building on granular computing (GrC), MAFRO first selects a minimal reduct of attributes, then constructs fuzzy granules with a Gaussian membership function to extract concise decision rules and realizes unlearning through the influence function. Specifically, instead of training with the full set of attributes, we use the reduct, a minimal subset of attributes that can classify the data with the same accuracy as the full set of attributes. Next, we extract fuzzy rules based on the reduct. Finally, fusing the generated rules establishes the linear model with strongly convex loss functions. In this way, MAFRO can quantify the divergence caused by attribute deleting and update the model without retraining it, thereby adapting the influence of data removal on the model and accelerating the unlearning process. We conduct extensive experiments to evaluate MAFRO on ten typical datasets in terms of performance and unlearning speed. We compare MAFRO with the state-of-theart algorithms. Experimental results demonstrate that MAFRO enhances accuracy by an average of 6.96%, and achieves up to 236× speedup for attribute unlearning tasks.

Sliding window aggregation, which extracts summaries from data streams, is a core operation in streaming analysis. Though existing sliding window algorithms that perform single eviction and insertion operations can achieve a worst-case time complexity of$O(1)$for in-order streams, real-world data streams often involve out-of-order data and exhibit burst data characteristics, which pose performance challenges to these sliding window algorithms. To address this challenging issue, we proposeGecko- a novel sliding window aggregation algorithm that supports bulk eviction. Gecko leverages a granular-based eviction strategy for various bulk sizes, enabling efficient bulk eviction while maintaining the performance close to that of in-order stream algorithms for single evictions. For large data bulks, Gecko performs coarse-grained eviction at the chunk level, followed by fine-grained eviction using leftward binary tree aggregation (LTA) as a complementary method. Moreover, Gecko partitions data based on chunks to prevent the impacts of out-of-order data on other chunks, thereby enabling efficient handling of out-of-order data streams. We conduct extensive experiments to evaluate the performance of Gecko. Experimental results demonstrate that Gecko exhibits superior performance over other solutions, which is consistent with theoretical expectations. In real-world data scenarios, Gecko improves the average throughput of the state-of-the-art algorithm b_FiBA by 1.7 times, with a maximum improvement of up to 3.5 times. Gecko also demonstrates the best latency performance among all compared schemes.

In programmable networks, measurement tasks are placed on programmable switches to monitor network traffic at line rate. These tasks typically require substantial resources (e.g., significant SRAM), while programmable switches are constrained by limited resources due to their hardware design (e.g., Tofino ASIC), making distributed deployment essentially. Measurement tasks must monitor specific network locations or traffic flows, introducing significant complexity in deployment optimization. This target-constrained nature makes task optimization on switches (e.g., task merging) become device-dependent and order-dependent, which can lead to deployment failures or performance degradation if ignored. In this paper, we introduceDisPLOY, a novel target-constrained distributed deployment framework specifically designed for network measurement tasks on the data plane.DisPLOYenables operators to specify monitoring targets—network traffic or device/link—across multiple switches. Given the monitoring targets,DisPLOYeffectively minimizes redundant operations and optimizes deployment to achieve both resource efficiency (e.g., minimizing stage consumption) and high-performance monitoring (e.g., high accuracy). We implement and evaluateDisPLOYthrough deployment on both P4 hardware switches (Intel Tofino ASIC) and BMv2. Experimental results show thatDisPLOYsignificantly reduces stage consumption by up to 66% and improves ARE by up to 78.4% in flow size estimation while maintaining end-to-end performance.

Service function chains (SFCs) are sequences of network functions that provide specific services to meet operators’ needs in today's ISPs and datacenter networks. To improve the performance of SFCs, programmable data planes are used to leverage their low latency and high performance packet processing. However, SFCs need to be adaptable to dynamics such as changes in requirements and attributes. Therefore, the ability to migrate SFCs is essential. Unfortunately, migrating SFCs in distributed programmable data planes is challenging due to the risk of degraded performance and failure to meet SFCs requirements and resource constraints in switches. In this paper, we proposeMonte, which provides an effective SFCs migration scheme in distributed programmable data planes. We build a novel integer programming model to represent the migration process with constraints on resource limitations of switches and SFCs attributes in the distributed data plane. Additionally, an SFCs migration algorithm is designed to optimize the migration cost by deeply analyzing resource allocation in the switch pipeline.Montehas been implemented on both P4 software switches (Bmv2) and hardware switches (Intel Tofino ASIC). Extensive evaluation results show that the migration cost inMonteis 94.03% lower on average than the state-of-the-art deployment scheme, andMontecan effectively save pipeline resources.

Programmable data plane (PDP) has emerged as a powerful platform for line-rate packet processing, utilizing on-chip register memory to execute stateful applications. Yet most existing efforts concentrate on static approaches for allocating register memory, necessitating switch restarting and service interruption. Despite the availability of research on sharing memory for concurrent applications, the rigid requirement of limiting memory sharing to the same pipeline stages hampers application flexibility and poses scalability challenges. To address this limitation, we presentFlxVRM,a flexible register memory virtualization layerfor data plane P4 programs which supports high-flexibility sharing of register memory for concurrent applications on PDP.FlxVRMenables memory allocation at any stage and location of the pipeline on PDP for each application at run time. To reduce resource usage during virtualization in the data plane pipeline,FlxVRMfurther merges different tables and actions with similar structures within P4 programs. Additionally,FlxVRMprovides a compiler to generate data plane programs for virtualization as well as the control plane API configuration. A prototype ofFlxVRMis implemented based on P4 hardware switches with Intel Tofino ASIC. Our experiment results show thatFlxVRMsignificantly improves the allocatable memory space for applications by up to 50%, while reducing the resource of the table up to 68%.

The reduction of energy consumption will be even more urgent in cloud data centers due to the explosive increase of application data. Virtual machine (VM) integration is a relatively standard technology currently applied for computing facilities of data centers. However, excessive VM consolidation can easily lead to local hot spots that lower the energy efficiency and reliability of data centers. In addition, on account of the impact of heat recirculation in data centers, the traditional VM scheduling strategy cannot comprehensively ponder optimizing the holistic data center energy, which encompasses both server energy and cooling energy. To handle these issues, we proposedEAVMS- an Energy-Aware VM Scheduling approach for minimizing the holistic energy consumption of data centers. EAVMS adopts a two-phase approach to gain energy efficiency while guaranteeing QoS. First, EAVMS leverages a Blended Genetic algorithm and Simulated Annealing algorithm (BGSA) to optimize the initial placement of VMs. Second, EAVMS utilizes a dynamic migration algorithm to achieve effective migration by setting a maximum server temperature threshold without violating the service level agreement (SLA) that cuts down energy consumption by moderating the hot spots of servers. We conducted extensive experiments using two real-world traces (i.e., PlanetLab and Google Cluster datasets) to evaluate the effectiveness of EAVMS. The experimental results unveil that our approach is capable of saving 3.23$ \%$–43.07$ \%$in the holistic energy consumption of cloud data centers with only a tiny service performance degradation compared to other state-of-the-art alternatives (e.g., MJPM, GRANITE, TAS, XINT-GA, and Random).

Escalating host temperatures in data centers can create hot spots, significantly boosting cooling costs and impacting reliability. Accurately predicting host temperatures is of the utmost importance in managing resources effectively. Existing temperature estimation solutions are inefficient due to a lack of accurate prediction. To this end, we proposed a deep learning-based thermal prediction approach called DLTPA, which aims to minimize the temperature and power consumption of virtual machines (VMs) in data centers. Specifically, we built a Long Short-Term Memory (LSTM) network model to accurately predict host temperature and power consumption, demonstrating superior performance over traditional algorithms. Our LSTM model exhibits exceptional accuracy in predicting thermal and energy dynamics, achieving an R2value of 0.98 in power consumption prediction, indicating exact forecasts. Furthermore, we design an efficient VM placement strategy to achieve host peak temperature reduction by rationally arranging VM tasks. The experimental results demonstrate that the DLTPA significantly improves over other leading-edge algorithms. It reduces peak power consumption by 3.64% to 9.39%, lowers average temperature by 3.21% to 7.96%, achieves a 0% SLA violation rate, and maintains a high level of load balancing at 19.8%.

In-network computing (INC) is a new paradigm that allows applications to be executed within the network, rather than on dedicated servers. Conventionally, INC applications have been exclusively deployed on the data plane (e.g., programmable ASICs), offering impressive performance capabilities. However, the data plane’s efficiency is hindered by limited resources, which can prevent a comprehensive deployment of applications. On the other hand, offloading compute tasks to the control plane, which is underpinned by general-purpose servers with ample resources, provides greater flexibility. However, this approach comes with the tradeoff of significantly reduced efficiency, especially when the system operates under heavy load. To simultaneously exploit the efficiency of data plane and the flexibility of control plane, we propose Carlo, a cross-plane collaborative optimization framework to support the network-wide deployment of multiple INC applications across both the control and data plane. Carlo first analyzes resource requirements of various INC applications across different planes. It then establishes mathematical models for resource allocation in cross-plane and automatically generates solutions using proposed algorithms. We have implemented the prototype of Carlo on Intel Tofino ASIC switches and DPDK. Experimental results demonstrate that Carlo can compute solutions in a short time while avoiding performance degradation caused by the deployment scheme.

Edge servers usually store collected data in cloud servers and use deduplication techniques to remove redundancy. However, edge servers can also perform deduplication during data collection. This requires transferring fingerprints from the cloud servers to the edge servers for assistance. Since the large volume of fingerprint data on the cloud server, for example, 1 PB of data corresponds to 8 TB of fingerprints, transferring all fingerprints to the edge servers is impractical. Therefore, we propose a fingerprint offloading strategy. Only a small amount of fingerprints and data chunks needed for edge deduplication are offloaded from the cloud server to the edge server, enabling cloud-edge collaborative deduplication. The general process is as follows: First, the edge server collect a large amount of data from various devices, divides the data into chunks, and calculates fingerprints to identify unique data chunks and fingerprints. Then, the edge server upload the fingerprints to the cloud server. The cloud server check the fingerprints and offload the data chunks corresponding to existing fingerprints back to the edge server. Upon receiving these data chunks, the edge server performs thorough deduplication. Finally, the deduplicated data chunks are uploaded to the cloud server, ensuring that only unique data is transmitted to the cloud server. Experiments used chunks ranging from 1 KB to 16 KB, with an average size of 4 KB, and employed three real backup datasets. The results showed that the edge server computation time was reduced by 48.1%, metadata storage was reduced by 98.1%, and the upload volume from the edge server to the cloud server decreased by 87.1%. The size of fingerprints and data chunks offloaded to the edge server ranged from 9.4 MB to 1624.2 MB.

Information granules are effective in revealing the structure of data. Therefore, it is a common practice in data mining to use information granules for classifying datasets. In the existing granular classifiers, the information granules are often classified according to the standard membership function only without considering the influence of different feature weights on the quality of granules and label classification results. In this article, we utilize the feature weighting of data to produce the information granules with high consistency rate called FIG. Firstly, we use consistency rate and contribution scores to generate information granules. Then, we propose a granular two-stage classifier GTC based on FIG. GTC divides the data into fuzzy and fixed points and then calculates the interval matching degree to assign data points to the most suitable cluster in the second step. Finally, we compare FIG with two state-of-the-art granular models (T-GrM and FGC-rule), and classification accuracy is also compared with other classification algorithms. The extensive experiments on synthetic datasets and public datasets from UCI show that FIG has sufficient performance to describe the data structure and excellent capability under the constructed granular classifier GTC. Compared with T-GrM and FGC-rule, the time overhead required for FIG to obtain information granules is reduced by an average of 51.07%, the per unit quality of the granules is also increased by more than 14.74%. Compared with other classification algorithms, an average of 5.04% improves GTC accuracy.

Virtual machine (VM) consolidation strategies are widely used in cloud data centers (CDC) to optimize resource utilization and reduce total energy consumption. Although existing strategies consider current and future resource utilization, the impact of sudden bursts in historical resource utilization on the hosts has been underestimated in uncertain future periods. Insufficient analysis of historical resource utilization may increase the risk of host overloading and Service Level Agreement Violation (SLAV). By defining historical and future trends based on resource utilization, we propose a novel combined trend VM consolidation (CTVMC) strategy which can effectively reduce energy consumption and SLAV. The VMs with the largest combined trend are selected for migration to prevent host overloading. Based on the temporal locality and prediction technique, CTVMC then employs the past, present, and future resource utilization to filter candidate hosts, and identifies the most complementary host to place VM using combined trends. We conduct extensive simulation experiments with PlanetLab Trace and Google Cluster Trace in the CloudSim simulator. Compared with the well-known strategies, CTVMC strategy using the PlanetLab Trace can reduce the number of migrations by over 72.39%, SLAV by over 75.85%, and ESV (a combined metric that judges the trade-off between energy consumption and SLAV) by over 81.54%. According to the Google Cluster Trace, our strategy can reduce the number of migrations by over 61.51%, SLAV by over 37.37%, and ESV by over 35.30%.

With high scalability and flexibility, serverless computing is becoming the most promising computing model. Existing serverless computing platforms initiate a container for each function invocation, which leads to a huge waste of computing resources. Our examinations reveal that (i) executing invocations concurrently within a single container can provide comparable performance to that provided by multiple containers (i.e., traditional approaches); (ii) redundant resources generated within a container result in memory resource waste, which prolongs the execution time of function invocations. Motivated by these insightful observations, we propose FaaSBatch - a serverless framework that reduces invocation latency and saves scarce computing resources. In particular, FaaSBatch first classifies concurrent function requests into different function groups according to the invocation information. Next, FaaSBatch batches the invocations of each group, aiming to minimize resource utilization. Then, FaaSBatch utilizes an inline parallel policy to map each group of batched invocations into a single container. Finally, FaaSBatch expands and executes invocations of containers in parallel. To further reduce invocation latency and resource utilization, within each container, FaaSBatch reuses redundant resources created during function execution. We conduct extensive experiments based on Azure traces to evaluate the effectiveness and performance of FaaSBatch. We compare FaaSBatch with three state-of-the-art schedulers Vanilla, SFS, and Kraken. Our experimental results show that FaaSBatch effectively and remarkably slashes invocation latency and resource overhead. For instance, when executing I/O functions, FaaSBatch cuts back the invocation latency of Vanilla, SFS, and Kraken by up to 72.58%, 74.10%, and 72.62%, respectively; FaaSBatch also slashes the resource overhead of Vanilla, SFS, and Kraken by 70.2% to 98.40%, 67.74% to 98.12%, and 43.01% to 78.90%, respectively.

Read-disturb problem plays a pivotal factor in the performance of NAND flash memory, because it deteriorates the read-disturb errors of NAND flash. Although ECC, read retry, and read reclaim technologies are designed to correct read-disturb errors, these techniques drastically increase read latency and degrade read performance. Moreover, modern SSDs implement a buffer in the built-in DRAM to store frequently accessed data, which can cache hot read data to alleviate the read-disturb problem. Unfortunately, the buffer primarily serves write requests to curtail write operations in flash memory, and ignores the ever-increasing requirement from users for read latency. To address this issue, we propose a read-disturb-aware dynamic buffer management scheme called Minato that reduces read-disturb errors with rationally read buffer management, aiming to improve the read performance of SSDs. Minato includes two distinctive and vital features. First, Minato dynamically adjusts the size of the read buffer and write buffer through the hit situation of requests, thus increasing the size of the read buffer while maintaining the write hit of the write buffer. Second, to further reduce read-disturb errors, Minato implements a read buffer filter to preferentially cache hot read data disturbing more valid pages into the read buffer. We compare Minato with two state-of-art schemes -BPLRU and GCaR in terms of write hit ratio, read-disturb counts, and read/write response time. The experimental results derived from nine real-world workload traces show that Minato efficiently alleviates the read-disturb problem of flash memory without affecting the write hit ratio, and significantly improves read/write performance. In particular, compared with the existing schemes, Minato slashes the read/write response time by an average of 34.6%.

Vehicle re-identification is a crucial task that matches images of the same vehicle across different camera viewpoints. Many previous attention-based studies have approached this problem by exploring the regions of interest in vehicles. However, the generated attention in these models is susceptible to noisy data, as they are unable to provide powerful supervision to distinguish biased and unbiased clues during the attention learning process. To address the problems mentioned above, we aim to design a robust vehicle re-ID network that utilizes the causal effect to effectively transfer attention from biased to unbiased clues. In this paper, we propose an unbiased feature-enhanced network (UnbiasNet), which consists of an unbiased feature-aware block (UFAB) and a novel causal effect-based joint constraint (CEC). In particular, we propose an unbiased feature-aware block as an attention module to extract rich and discriminative information. We conduct a counterfactual intervention on our attention module to generate biased feature representations. Moreover, we propose a novel causal effect-based joint constraint that consists of original prediction constraint and total indirect effect constraint. The original prediction constraint ensures that unbiased feature-aware block converges correctly. The total indirect effect constraint utilizes the generated biased features as supervisory information to motivate unbiased feature-aware block to explore a greater number of unbiased features during the training process. Our approach had an inference time of 1.39 ms per image, which introduces only a few parameters during the training phase and none during the testing phase. We carry out comprehensive experiments to illustrate the effectiveness of the UnbiasNet on three challenging datasets.

The expansion of data centers, driven by the continuous development of network services, has led to a significant issue of high energy consumption. Due to the real-time interaction between IT and non-IT equipments, it is difficult to consider the holistic energy consumption of heterogeneous data centers. Therefore, this paper proposes a holistic-energy-aware-virtual machine placement (VMP) strategy for data centers that use heterogeneous resources to provide services. Firstly, we propose the energy-aware VMP strategy by using the probabilistically determining mechanism to effectively minimize the number of activated servers and improve server resource utilization. Secondly, within this strategy, we leverage dynamic voltage and frequency scaling (DVFS) technology, enabling nodes to operate at lower frequencies and voltages while meeting performance requirements, thus further reducing computing energy consumption. Thirdly, in addition to the previous two points, the probabilistic determined genetic algorithm (PDGA) is proposed to reasonably distribute the workloads based on the heat-recirculation effect and reduces the cooling energy consumption. The above mechanisms collectively optimize the global energy consumption of heterogeneous data centers. Experimental results demonstrate that the PDGA can significantly reduce the energy consumption of IT and non-IT equipment. The total energy consumption of the data center is significantly reduced (the PDGA is 20.83% lower than the simulated annealing based algorithm and 20.76% lower than the big data task scheduling algorithm based on thermal-aware and DVFS-enabled techniques).

Cloud computing has emerged as an infrastructure in the era of digital economy and has been widely applied in various fields. Virtual Machine(VM) management is the key mechanism in a Cloud Data Center(CDC). A typical VM management system is responsible for VM allocation and VM reallocation, and it is usually designed to optimize specific objectives, especially for the metrics of energy consumption, resource wastage, communication cost, and Service Level Agreement Violations (SLAV). However, it is greatly challenging to optimize these metrics at the same time, and most existing VM management strategies focus on optimizing part of the above four metrics. In this paper, we propose a Holistic Virtual Machine Management (HVMM) strategy to optimize the energy consumption, resource wastage, communication cost, and SLAV simultaneously. First, we define two parameters, the Compatibility and the Performance-to-Power Ratio (PPR), for VM allocation to optimize energy consumption and resource wastage. Then, we propose a reallocation approach based on spectral clustering that can handle dynamic traffic between VMs without a priori knowledge of the traffic between VMs, and it takes a slight expense of resource wastage and energy consumption to reduce communication cost between VMs and ensures low SLAV risk. To evaluate the performance of HVMM, we compared the proposed strategy with the state-of-the-art strategies in various experiments on real-world traces. Compared with the other strategies, the resource wastage of HVMM is reduced by 62%. Simultaneously, the communication cost is reduced by about 26%, the energy consumption is reduced by about 5%, and SLAV is much lower than that of the others.

In elastic cloud service systems, it is a challenge to evaluate and match the fluctuating resource demand of workloads. Existing studies typically monitor workload characteristics and build models that map these characteristics to actual demand. However, workload characteristics are multidimensional, and the impact of each dimension on resource demand differs, so it requires differentiated treatment when building models. This paper proposes a Correlation-Based Universal Resource Matching (CBURM) strategy to realize a Heterogeneous and Elastic Cloud Service System (HECSS). CBURM consists of a Correlation-based resource Demand Evaluation (CDE) method and a Universal Resource Measurement (URM) scheme. Specifically, CDE discriminates the relevance of each dimension in workload characteristics, based on the correlations between workload characteristics and the demand. Then, it generates resource demand decisions dimension by dimension, from the most relevant to the least relevant dimensions. After that, it generates a complete decision tree model to evaluate subsequent workload demand for heterogeneous resources. Finally, URM optimizes the resource allocation to achieve a low-overhead resource matching. Experimental results show that, URM reduces the total comprehensive operation cost by 82%+, compared to a normal resource allocation scheme. Additionally, CDE outperforms two state-of-the-art methods (LTP and 2SP), with its performance closer to the ideal baseline. Specifically, CDE achieves a 40.275% overall resource saving rate, which is 38.62% higher than LTP and 8.46% higher than 2SP. Besides, CDE achieves a 92.43% average service quality satisfaction ratio, higher than the 82.9% and 88.83% achieved respectively by LTP and 2SP.

With the rapid growth of cloud computing, frequent workload bursts show an increasing influence on the Quality of Service (QoS) and energy efficiency of cloud-based data centers. Existing virtual machine placement schemes are expected to optimize either QoS or energy efficiency for cloud data centers running under bursty workload conditions. To bridge this gap, we propose a burst-aware and thermal-efficient virtual machine placement technique calledBTVMP. BTVMP adopts a two-step strategy to achieve energy efficiency while assuring QoS. First, BTVMP leverages a split-and-recombine algorithm – SAR – to deal with bursty workloads. SAR prioritizes critical workloads while preventing low-priority workloads from starvation, thereby assuring QoS. Second, BTVMP utilizes an enhanced simulated annealing algorithm calledESAto offer optimal thermal-efficient virtual machine placement (VMP) solutions, aiming to minimize the energy consumption of data centers. To facilitate estimating energy consumption, we integrate into BTVMP a thermal model that takes into account heat re-circulation effects. We conduct extensive experiments with a real-world trace. We compare BTVMP with the leading-edge VMP strategies, including Genetic Algorithm (XINT-GA), Power-Aware and Performance-Guaranteed Virtual Machine Placement (PPVMP), Peak Load Scheduling Control Method (PLSC), First Come First Serve (FCFS), and GReedy based scheduling Algorithm miNImizing Total Energy (GRANITE). The experimental results unveil that BTVMP not only enhances QoS but also exhibits superb energy efficiency. In particular, BTVMP reduces PLSC's workload delay and FCFS's critical workload delay by 18$\%$and 11$\%$, respectively. Moreover, BTVMP lowers the total energy consumption of the three alternative algorithms –GRANITE, XINTGA, PPVMP, and PLSC – by anywhere between 27.8$\%$and 49.4$\%$.

Network measurement is essential for efficient network management and operations. In-band network telemetry (INT) offers fine-grained per-device per-packet information which could provide full-visibility for networks. However, the existing solutions fall short in achieving high accuracy, generality, and low overhead simultaneously. To address this limitation, we introduceOffsetINTto meet these three criteria. The key idea ofOffsetINTis to use minimal bits to carry collected states during monitoring, which is based on our observation that the value of telemetry states are usually very close (e.g., the time of adjacent arrival packets) or small (e.g., only a few tens of microseconds for processing latency) for most of the time in real networks. Instead of embedding complete values of state in packets,OffsetINToptimizes bit usage by encoding an offset (using fewer bits), which is carried in-band by passing packets to the end-hosts for recovery and analysis. We theoretically derive the bounds of bandwidth mitigation forOffsetINT. We have implementedOffsetINTin both P4 hardware switches (with Intel Tofino ASIC) and BMv2. Expensive evaluation results show thatOffsetINTcan achieve an accuracy of up to 100% compared to the original INT while reducing INT bandwidth by up to 48%.

The explosion of large-scale data has increased the scale and capacity of storage clusters in data centers, leading to huge power consumption issues. Cloud providers can effectively promote the energy efficiency of data centers by employing energy-aware data placement techniques, which primarily encompass storage cluster's power and cooling power. Traditional data placement approaches do not diminish the overall power consumption of the data center due to the heat recirculation effect between storage nodes. To fill this gap, we build an elaborate thermal-aware data center model. Then we propose two energy-efficient thermal-aware data placement strategies, ETDP-I and ETDP-II, to reduce the overall power consumption of the data center. The principle of our proposed algorithm is to utilize a greedy algorithm to calculate the optimal disk sequence at the minimum total power of the data center and then place the data into the optimal disk sequence. We implement these two strategies in a cloud computing simulation platform based on CloudSim. Experimental results unveil that ETDA-I and ETDP-II outperform MinTin-G and MinTout-G in terms of the supplied temperature of CRAC, storage nodes power, cooling cost, and total power consumption of the data center. In particular, ETDP-I and ETDP-II algorithms can save about 9.46%-38.93% of the overall power consumption compared to MinTout-G and MinTin-G algorithms.

With high scalability and flexibility, serverless computing is becoming the most promising computing model. Existing serverless computing platforms initiate a container for each function invocation, which leads to a huge waste of computing resources. Our examinations reveal that (i) executing invocations concurrently within a single container can provide comparable performance to that provided by multiple containers (i.e., traditional approaches); (ii) redundant resources generated within a container result in memory resource waste, which prolongs the execution time of function invocations. Motivated by these insightful observations, we propose FaaSBatch - a serverless framework that reduces invocation latency and saves scarce computing resources. In particular, FaaSBatch first classifies concurrent function requests into different function groups according to the invocation information. Next, FaaSBatch batches the invocations of each group, aiming to minimize resource utilization. Then, FaaSBatch utilizes an inline parallel policy to map each group of batched invocations into a single container. Finally, FaaSBatch expands and executes invocations of containers in parallel. To further reduce invocation latency and resource utilization, within each container, FaaSBatch reuses redundant resources created during function execution. We conduct extensive experiments based on Azure traces to evaluate the effectiveness and performance of FaaSBatch. We compare FaaSBatch with three state-of-the-art schedulers Vanilla, SFS, and Kraken. Our experimental results show that FaaSBatch effectively and remarkably slashes invocation latency and resource overhead. For instance, when executing I/O functions, FaaSBatch cuts back the invocation latency of Vanilla, SFS, and Kraken by up to 92.18%, 89.54%, and 90.65%, respectively; FaaSBatch also slashes the resource overhead of Vanilla, SFS, and Kraken by 58.89% to 94.77%, 43.72% to 90.39%, and 42.99% to 78.88%, respectively.

Information granule is a classic mathematical paradigm in the field of data mining. Existing research focuses on improving granule quality to optimize models. However, in these studies, they do not consider that the fairness of the granular model will affect the performance of the granule, especially in some severe social issues (Law, Finance, Education, and more), often due to the participation of sensitive features, the application results of the granule (such as classification) cause population bias. Thus, we construct an adaptive probability replacement resampling model (APR) in Pre-Processors to reduce the bias of sensitive features on granular results. Then, in In-Processors, we add fairness optimizations (FO) to improve the fairness of information granules. Finally, we propose a fair information granule model based on adaptive probability replacement resampling called APRG(APR+FO). We select three loan datasets to verify the feasibility of the granular model according to the current sensitive lending issues in the financial field. The experimental results show that compared with the existing granular models, our proposed method dramatically improves the fairness of the constructed granular models, especially the APRG model, which has an average increase of 46.6% on Demographic Parity (DP) and 77.9% on Equalized Odds (EO). Compared with other existing granular models, the granule quality of APRG is increased by 38.88% on average; in terms of classification accuracy, the average decrease is only 2.03%.

Vehicle Re-identification (Re-ID) is very important in intelligent transportation and video surveillance. Most previous works mainly extract discriminative features from the visual appearance of vehicles. However, the irrelevant background in the image will affect the extracted features, and it is also important to extract fine-grained features because the vehicles have the same style. Previous works extracted robust features by manual annotation, which was inefficient. To solve this problem, this paper proposes a self-guided feature mining network (SFMNet) that can eliminate background intervention without annotation, while mining fine-grained feature information of vehicles. In order to eliminate background intervention and mine fine-grained features without resorting to annotations, we have carefully designed two novel modules. (i) The noise patch filter (NPF) module can identify the image background without annotation, and filter out the background to eliminate the intervention of the background on the image features. (ii) The salient feature extraction (SFE) module uses self-attention as a guide without vehicle component annotation to mine fine-grained features and enhance discriminative visual cue features. Extensive experiments demonstrate the effectiveness of our method, and we achieve state-of-the-art results on three publicly available datasets, including Veri-776, VehicleID, and VERI-WILD.

Increasing workload conditions lead to a significant surge in power consumption and computing node failures in data centers. The existing workload distribution strategies focused on either thermal awareness or failure mitigation, overlooking the impact of node failures on the energy efficiency of cloud data centers. To address this issue, a new holistic model is built to characterize the impacts of workloads, computing and cooling costs, heat recirculation, and node failure on the energy efficiency of cloud data centers. Leveraging such a holistic model, we propose a novel thermal-aware workload distribution strategy calledHGSAthat takes node failure into accountand can improve the energy efficiency of cloud data centers. Our empirical findings confirm that (i) faulty nodes lead to a large rise in power consumption, and (ii) failure locations play a vital role in the power consumption of data centers. Experimental results unveil that HGSA is adroit at making near-optimal decisions in workload distribution strategies. In particular, HGSA cuts down the minimum inlet temperature by 5.2$\%$-15$\%$, improves the maximum air temperature of a Computer Room Air Conditioner (CRAC) model by 4.2$\%$-26.5$\%$, lowers the cooling cost by 15.4$\%$-50$\%$compared to the existing solutions. Furthermore, HGSA cuts back the total power consumption by 0.65$\%$-78$\%$.

As cloud platforms become increasingly popular, accurately understanding I/O behaviors in modern cloud storage is of paramount importance for system design and optimization. This paper sheds new light on the correlation of inter-arrival times of both read and write requests at the block level in four representative cloud storage workloads – AliCloud, Systor’17, MSRC and FIU. Our study reveals that I/O arrivals at the block level are very complex in modern cloud storage. There is a certain degree of correlation in the long-term timescale for request arrival intervals in AliCloud and Systor’17_read. Request arrival intervals in MSRC, FIU and Systor’17_write, however, are almost uncorrelated. The Gaussianity test confirms that I/O burstiness appears to be Gaussian in AliCloud_write and Systor’17_read, but the burstiness is non-Gaussian in other workloads. Importantly, we unfold the existence of self-similarity in cloud storage workloads with a certain degree of correlations, via visual evidence, the autocorrelation structure of the aggregated process of I/O request sequences, and Hurst parameter estimates. We further design an alpha-stable workload model for synthetic I/O generation, and the experimental results demonstrate that our model has an edge over conventional models in terms of accurately emulating I/O burstiness.

Multibit 3-D NAND flash-based solid-state disks (SSDs), offering high storage density, contain multiple types of pages to accommodate multiple bits per physical cell. Full sequence program or FSP can program multiple pages in a word line at a time, thereby improving write throughput. Unfortunately, large-grained FSP operations coarsely aggregate consecutive logical pages on the same word line, which adversely affects the parallelism and latency of read requests. Moreover, FSP smooths the program latencies for different types of pages, whereas the pages still exhibit various read latencies. Multiple read latencies and lower read parallelism noticeably deteriorate the completion efficiency of read requests: SSD performance is degraded. To address this issue, we propose an FSP data allocation scheme called FSPDA that incorporates the physical structure characteristics of multibit 3-D NAND, aiming to bolster the read performance of 3-D NAND Flash-based SSDs. FSPDA embraces two distinctive and vital features. First, according to the distance between logical pages, FSPDA allocates logical pages to specified parallel units and stipulates that consecutive logical pages must be assigned to different planes, thus improving read parallelism and data locality. Second, to further reduce read latency, FSPDA employs cache hits to determine hot and cold data to be placed to low-latency and high-latency pages, respectively. We compare FSPDA with two state-of-the-art schemes—OSPADA and single-operation-multiple-location—in terms of multiplane read (MPR) counts, read response time, and GC counts under eight real-world workloads. The experimental results show that compared with the existing schemes, FSPDA slashes the number of MPR counts, read response time, and the number of GC counts by an average of 34.4%, 28.5%, and 13.6%, respectively.

Garbage collection or GC running in the controller of 3-D NAND flash-based solid-state disks—SSDs—plays a critical role in the performance of storage systems. SSD manufacturers have developed various GC solutions based on internal data movement or IDM to mitigate the impacts of GC on request latency. Due to the circuit characteristics of flash memory, the existing IDM-based GC strategies are restricted by page parity during data movement: odd pages must be migrated to odd pages, and even pages to even pages. When migrating two consecutive pages with the same parity, the free page between the two migrated pages will be wasted after the migration is complete. This ever-increasing page waste problem inevitably deteriorates the storage space utilization of flash memory, thereby degrading the overall performance of 3-D NAND flash-based SSDs. To address this issue, we propose a parity-check GC scheme called PcGC to revamp SSD performance by alleviating page waste during GC. We build a parity-check unit in PcGC to facilitate checking the parity of migrated valid pages and destination pages. According to the parity results offered by the parity-check unit, PcGC dynamically adjusts the migration order of valid pages during the course of GC. In doing so, PcGC fundamentally averts page waste caused by the page parity restriction, thereby enhancing 3-D NAND flash performance. We quantitatively evaluate the performance of PcGC in terms of wasted pages, storage utilization, GC counts, write amplification, and average response time. We compare PcGC against the two state-of-the-art schemes—Amphibian and Tiny-tail flash (TTflash). The experimental results derived from the nine real-world workload traces unfold that compared with Amphibian and TTflash: 1) PcGC curtails the number of wasted pages by up to 91.4% with an average of 53.75%; 2) cuts back the number of GC counts by up to 52.2% with an average of 11.9%; and 3) slashes average write response time by up to 77.8% with an average of 13.0%.

A large number of read-disturb-induced rewrites are performed in the background [also known as Read Reclaim (RR)] to alleviate the read-disturb issue in NAND flash memory-based SSDs. RR can significantly degrade the performance and shorten the service life of SSD in read-intensive workloads. To address this issue, we propose a novel read-disturb management approach called Cocktail that mixes a small proportion of hot-read pages with a large proportion of cold-read pages, thereby avoiding clustering hot-read pages into a few blocks. Motivated by the insight that RR operations are frequently triggered by hot read-pages, Cocktail first prefills a portion of each block with cold data extracted from user requests. Then, Cocktail fills the prefilled blocks with write-back data caused by RR to create read-balanced blocks. We integrate two thresholds, write pool capacity and the ratio of RR-write data to User-write data, into Cocktail to govern the ratio of write-back data caused by RR to data of user requests in a block. Cocktail dynamically adjusts the two thresholds according to the characteristics of RR. Cocktail is conducive to decentralizing hot write-back data caused by RR across a broad range of blocks, thereby reducing the occurrence of second-time RR and the number of overall block reads. We compare Cocktail with three existing schemes baseline, redFTL, and IPR in terms of SSD service life, SSD response time, write amplification, and the number of garbage collections (GCs) under ten real-world workload conditions. Experimental results show that compared with the existing schemes, Cocktail reduces the number of RRs, the average response time, the 99-percentile tail latency, and the number of GCs by an average of 40.77%, 10.82%, 5.40%, and 12.29%, respectively. Cocktail also alleviates the write amplification of the three alternative schemes by an average of 49.57%.

Task offloading has been widely used to extend the battery life of intelligent mobile devices. Existing task offloading approaches, focusing on perfecting the balance between latency and energy consumption, completely ignore the impacts of the user preference caused by low battery anxiety. The existence of low battery anxiety - mobile users’ common fear of losing battery energy, especially when the battery energy is already low - causes users to trade high latency for prolonged battery life. Taking into account the user preference impacts on task offloading, we propose a novel offloading approach calledUPOAto obtain refined offloading policies between low latency and energy consumption based on user preferences. In UPOA, we start the study by defining a user preference rule that determines users’ offloading preferences according to battery energy status. Then, we build a fine-grained task offloading model to delineate the task distribution characteristics of each node in its offloading link. Guided by this model, we develop a task prediction algorithm based on the long-short-term-memory neural network model to provide task predictions that facilitate offloading policies. Lastly, we implement a particle-swarm-optimization-based online offloading algorithm. The offloading algorithm provides the best long-term offloading policies by incorporating the user preference determined by our user preference rule and the task predictions generated by our task prediction algorithm. To quantitatively evaluate the performance of UPOA, we conduct extensive experiments in a real-world cloud-edge environment. We compare UPOA with three state-of-the-art offloading approaches, DRA, DRL-E2D, and MUDRL under various conditions. Experimental results demonstrate that UPOA can make effective policies based on user preferences compared with the existing approaches. UPOA reduces average latency by 12.49% when battery energy is sufficient and extends battery life by 20.14% when battery energy is low.

Data deduplication has been broadly used in Cloud due to its storage space saving ability. An issue of deduplication is the contiguous data chunks in a segment may be scattered in different containers. This phenomenon is called data fragmentation. Because of data fragmentation, a restore process must reference various containers across a wide variety of segments, thereby hurting the restore performance. Capping methods that rewrite the data chunks of low Container Reference Ratio (CRR) containers are developed to alleviate data fragmentation. We analyze and observe from real traces that a number of segments only point to lowCRRcontainers, while some others only contain highCRRcontainers. This interesting observation is ignored by the existing capping methods which sort containers from a single segment, falling short in searching multiple segments collectively. Thus, the reference count of selected containers in the existing capping methods is still high. To address this problem, we propose a multi-segment greedy rewriting method named MGRM. MGRM sorts containers of segments in a sequential way. More specifically, given thei-thsegment currently being processed, MGRM will sort all the containers in the topi-thsegments. This salient searching feature enables MGRM to select and rewrite the true low-reference container set. Moreover, to achieve a good balance between deduplication ratio and restore performance, MGRM has two working modes: an optimal rewriting mode and a radical rewriting mode. When working in the optimal rewriting mode, MGRM aims to improve the deduplication ratio; when the radical rewriting mode, MGRM strives to improve the restore performance. MGRM adaptively switches the working mode according to workload. Furthermore, unlike the existing capping methods that improve restore performance at the cost of the deduplication ratio, MGRM pays attention to both aspects. Our extensive experimental results show that MGRM achieves high restore performance, coupled with a high deduplication ratio. In particular, compared with the two state-of-art schemes FC and FLC, MGRM improves the deduplication ratio and restore performance by up to 114.83% and 99.34%, respectively.

Fuzzy C-means clustering (FCM) has been a commonly used algorithm in fuzzy clustering for decades. However, it still faces two problems: how to determine the initial cluster center and how to determine the number of clusters. The recently proposed robust learning fuzzy C-means (RL-FCM) can automatically obtain the optimal number of clusters. However, it assumes that the initial cluster center is the entire dataset, which incurs a significant time cost and involves parameters that are also difficult to determine. Additionally, RL-FCM is unable to handle imbalanced datasets and datasets with a large span of sample attributes. Therefore, we propose a robust learning membership scaling fuzzy C-means algorithm based on new belief peaks (RL-MFCM). Within the framework of the confidence function, the neighbors of the sample points provide evidence for the sample points being cluster centers. Consequently, according to Jiang's combination rule, we consider the new belief peak as the initial cluster center. To avoid excessive interference of the mixing ratio of the cluster to the calculation of membership degree, we employ triangle inequality to improve the influence of the samples in the cluster in the clustering process. We analyze the time complexity of the proposed algorithm and conduct comparative experiments with existing fuzzy clustering algorithms on artificial and real datasets in the article. Experiments demonstrate that our proposed algorithm accurately estimates the number of clusters and exhibits superior clustering performance without needing initialization.

With the mushrooming growth of data volumes, data replica placement plays a key role in promoting the energy efficiency and Quality-of-Service (QoS) of data-intensive data centers. The existing data placement strategies mainly focus on storage performance improvement or QoS enhancement in data centers, but ignore the indispensable factor - heat recirculation. To bridge this gap, we propose a thermal-aware data replica placement strategy called TADRP, aiming to improve cooling efficiency and minimize the total power consumption of data-intensive data centers. TADRP leverages an ant colony optimization (ACO) algorithm coupled with Laplacian probability distribution to find a quasi-optimal disk sequence (or Disk Sequence for short), which consists of disks selected from different rack servers to place data replicas. TADRP categorizes disks of Disk Sequence into active and inactive ones, by placing hot and cold replicas on active and inactive disks, respectively. We quantitatively evaluate TADRP in terms of cooling costs, total power consumption, number of power-state transitions, and execution time. We compare TADRP with four alternative solutions, namely, Random, Hadoop, SRS, and CDP-NSGAII. Experimental results show that TADRP can reduce the cooling costs and the total power of the existing solutions by 14.7% - 61.7% and 19.2%-55.1%, respectively, without undesirable I/O performance drops.

Inline deduplication removes redundant data in real-time as data is being sent to the storage system. However, it causes data fragmentation: logically consecutive chunks are physically scattered across various containers after data deduplication. Many rewrite algorithms aim to alleviate the performance degradation due to fragmentation by rewriting fragmented duplicate chunks as unique chunks into new containers. Unfortunately, these algorithms determine whether a chunk is fragmented based on a simple pre-set fixed value, ignoring the variance of data characteristics between data segments. Accordingly, when backups are restored, they often fail to select an appropriate set of old containers for rewrite, generating a substantial number of invalid chunks in retrieved containers. To address this issue, we propose an inline deduplication approach for storage systems, called InDe , which uses a greedy algorithm to detect valid container utilization and dynamically adjusts the number of old container references in each segment. InDe fully leverages the distribution of duplicated chunks to improve the restore performance while maintaining high backup performance. We define an effectiveness metric, valid container referenced counts (VCRC) , to identify appropriate containers for the rewrite. We design a rewrite algorithm F-greedy that detects valid container utilization to rewrite low-VCRC containers. According to the VCRC distribution of containers, F-greedy dynamically adjusts the number of old container references to only share duplicate chunks with high-utilization containers for each segment, thereby improving the restore speed. To take full advantage of the above features, we further propose another rewrite algorithm called F-greedy+ based on adaptive interval detection of valid container utilization. F-greedy+ makes a more accurate estimation of the valid utilization of old containers by detecting trends of VCRC’s change in two directions and selecting referenced containers in the global scope. We quantitatively evaluate InDe using three real-world backup workloads. The experimental results show that compared with two state-of-the-art algorithms (Capping and SMR), our scheme improves the restore speed by 1.3×–2.4× while achieving almost the same backup performance.

Garbage collection (GC) plays a pivotal role in the performance of 3D NAND flash memory, where Copyback has been widely used to accelerate valid page migration during GC. Unfortunately, copyback is constrained by the parity symmetry issue: data read from an odd/even page must be written to an odd/even page. After migrating two odd/even consecutive pages, a free page between the two migrated pages will be wasted. Such wasted pages noticeably lower free space on flash memory and cause extra GCs, thereby degrading solid-state-disk (SSD) performance. To address this problem, we propose a page-state-aware cache scheme called PSA-Cache , which prevents page waste to boost the performance of NAND Flash-based SSDs. To facilitate making write-back scheduling decisions, PSA-Cache regulates write-back priorities for cached pages according to the state of pages in victim blocks. With high write-back-priority pages written back to flash chips, PSA-Cache effectively fends off page waste by breaking odd/even consecutive pages in subsequent garbage collections. We quantitatively evaluate the performance of PSA-Cache in terms of the number of wasted pages, the number of GCs, and response time. We compare PSA-Cache with two state-of-the-art schemes, GCaR and TTflash, in addition to a baseline scheme LRU. The experimental results unveil that PSA-Cache outperforms the existing schemes. In particular, PSA-Cache curtails the number of wasted pages of GCaR and TTflash by 25.7% and 62.1%, respectively. PSA-Cache immensely cuts back the number of GC counts by up to 78.7% with an average of 49.6%. Furthermore, PSA-Cache slashes the average write response time by up to 85.4% with an average of 30.05%.

Serverless computing is a leading force behind deploying and managing software in cloud computing. One inherent challenge in serverless computing is the increased overall latency due to duplicate computations. Our initial investigation into the function invocations of serverless applications reveals an abundance of duplicate invocations. Inspired by this critical observation, we introduceHashCache, a system designed to cache duplicate function invocations, thereby mitigating duplicate computations. In HashCache, serverless functions are classified into three categories, namely, computational functions, stateful functions, and environment-related functions. On the grounds of such a function classification, HashCache associates the stateful functions and their states to build an adaptive synchronization mechanism. With this support, HashCache exploits the cached results of computational and stateful functions to serve upcoming invocation requests to the same functions, thereby reducing duplicate computations. Moreover, HashCache stores remote files probed by stateful functions into a local cache layer, which further curtails invocation latency. We implement HashCache within theApache OpenWhiskto forge a cache-enabled serverless computing platform. We conduct extensive experiments to quantitatively evaluate the performance of HashCache in terms of invocation latency and resource utilization. We compare HashCache against two state-of-the-art approaches -FaaSCacheandOpenWhisk. The experimental results unveil that our HashCache remarkably reduces invocation latency and resource overhead. More specifically, HashCache curbs the 99-tail latency of FaaSCache and OpenWhisk by up to 91.37% and 95.96% in real-world serverless applications. HashCache also slashes the resource utilization of FaaSCache and OpenWhisk by up to 31.62% and 35.51%, respectively.

In this paper, we propose a thermal-aware data replica placement strategy called TADRP. This strategy is designed in two steps. First, the ant colony optimization (ACO) algorithm based on the Laplacian probability distribution obtains the near-optimal disk sequence with the minimum overall power consumption. Second, the near-optimal disk sequence is partitioned into the area of active and inactive disks; then, the sequence-based data placement policy places data replicas in the partitioned disk areas. Our objection is to adopt the TADRP strategy to improve cooling efficiency and reduce the overall power consumption of DDCs. To evaluate the overall power consumption of DDCs, we integrate TADRP into a thermal model that takes into account heat recirculation effects account. We apply a real dataset with different read/write ratios following the Zipf distribution to verify the effectiveness of TADRP for energy savings. Experiment results unveil that our TADRP is capable of offering about 19.2 %-55.1% for total energy savings without significantly degrading I/O performance against Random, Hadoop, SRS, CDP-NSGAIIIR schemes.

The development of cloud computing leads to an exponential growth of data, which requires expanding the storage capacity to meet the storage needs, in exchange the energy consumption of the data center will also increase. Many traditional data placement schemes attempt to achieve energy consumption minimization by optimizing the distribution of data requests, however, ignoring the impact of heat recirculation in data placement. To fill this gap, we propose a heat-recirculation-aware data placement strategy called HRADP to achieve optimized data placement to data centers. Furthermore, our strategy can minimize the overall energy consumption of data centers and improve throughput. Specifically, we consider heat recirculation between data nodes by regulating the energy consumption limit of each data node. We implement this data placement strategy on a data center simulation platform, CloudSim, and compare it with two data placement strategies, TIGER and HDFS. The experimental results unveil that HRADP achieves 2.6x - 25.7x performance improvement in the same energy consumption.

As customers take virtual machines (VMs) as their demands, high-efficient placement of VMs is required to reduce the energy consumption in data centers. Existing Virtual Machine placement (VMP) strategies can minimize energy consumption of data centers by optimizing resource allocation in terms of multiple physical resources (e.g., memory, bandwidth, CPU, etc.). However, these strategies ignore the role of heat recirculation in the data center, which can cause a huge energy waste in cooling. To address this problem, we propose a heat-recirculation-aware VMP strategy for reducing the energy consumption of data centers. This novel VMP strategy takes into account heat recirculation coupled with multiple physical resource allocation to reduce the energy consumption of data centers. We design a simulated annealing based algorithm called SABA to lower the energy consumption of data centers where multiple VMs are deployed. SABA remarkably cuts down the energy consumption of physical resources through two salient features. First, it obtains an approximation of the optimum with much fewer iterations than simulated annealing algorithm (SA). Second, it reduces the number of activated servers required for VM tasks. We quantitatively evaluate the performance of SABA in terms of algorithm efficiency, the number of activated servers and the energy-saving. We compare the performance of SABA with state-of-art XINT-GA, PPVMP, TSTD and SA algorithms. Moreover, we evaluate the efficiency of SABA by leveraging a real-world 50 hours trace from practical IBM cloud data centers. Experimental results indicate that our heat-recirculation-aware VM placement strategy provides a powerful solution for improving the energy efficiency of data centers (SABA improves energy efficiency of cooling by up to 13.2% over TSTD, 13.8% over XINT-GA and 45% over PPVMP algorithm).

In modern datacenters, huge energy consumption is a significant problem that remains to be solved. Previous works reduce the system energy consumption by switching the idle servers to a low-power state. However, the workload demands on servers change dynamically and mainly depend on the real-time workload status. To maintain the system energy efficiency, when providing servers with some servers reserved, the status of workloads should be carefully considered. Generally, the status of workloads is characterized by some key factors sampled from the workloads. However, under dynamic workload demands, the accuracy of these sampled values varies. To accurately assess the status of server workloads, in this paper, we propose a Dynamic Time Scale based server Provision (DTSP) method, which takes the variability of workloads into consideration when providing servers for workload demands. To obtain accurate factor values indicating real-time workload, DTSP samples several key workload factors, including the coefficient of variation of arrival intervals, the request arrival rates of current workload and previous workload, and the mean service time of current requests, with a dynamic compatible rate. With these sampled factors, DTSP can accurately estimate the demands of workloads on servers and provide appropriate numbers of servers for the dynamic workloads. Extensive experiments demonstrate that taking the workload variability into consideration, DTSP can significantly promote the energy efficiency of a datacenter.

Instantiated containers of an application are distributed across multiple Physical Machines (PMs) to achieve high parallel performance. Container placement plays a vital role in network traffic and the performance of containerized data centers. Existing container placement techniques are inadequate due to the ignorance of container traffic patterns. To solve this issue, we first investigate the network traffic between containers and observe that it exhibits a Zipf-like distribution. Motivated by this finding, we propose a novel container placement approach-Blender-by taking into account the Zipf-like distribution. Blender employs two algorithms calledRefineAlgandSplitAlgto divide containers of applications into blocks, and place these blocks across Virtual Machines (VMs). Blender exhibits two salient features: (i) it minimizes inter-block traffic by arranging the containers that communicate frequently in the same block. (ii) it achieves good load balancing by combining complementary blocks that request different resource types (e.g.,CPU-intensiveandmemory-intensiveblocks) and distributing these blocks across multiple VMs. The experimental results show that Blender significantly reduces communication traffic and network latency. In particular, Blender reduces the traffic of SBP and CA-WFD by 22% and 32%, respectively. Blender decreases network latency by 16% and 26% compared to SBP and CA-WFD. Furthermore, with Blender in place, the physical resources of hosting PMs are well balanced and utilized.

Instantiated containers of an application are distributed across multiple Physical Machines (PMs) to achieve high parallel performance. Container placement plays a vital role in network traffic and the performance of containerized data centers. Existing container placement techniques do not consider the container traffic pattern, which is inadequate. To resolve this conflict, we investigate network traffic between containers and observe that it exhibits a Zipf-like distribution. We propose a novel container placement approach - Blender - by leveraging the Zipf-like distribution. Based on network traffic correlation, Blender employs RefineAlg and SplitAlg to divide containers of applications into blocks, and place these blocks across virtual machines. Blender exhibits two salient features: (i) it minimizes inter-block traffic by arranging the containers that communicate frequently in the same block. (ii) it achieves good load balancing by combining blocks according to the resource types they require and distributing them across multiple PMs. We compare Blender against two state-of-the-art methods SBP and CA-WFD. The experimental results show that Blender significantly reduces communication traffic. In particular, for the same number of PMs, Blender reduces the traffic of SBP and CA-WFD by 22% and 32%, respectively. Furthermore, with Blender in place, the physical resources of hosting PMs are well balanced and utilized.

Data deduplication has been widely used to improve storage space utilization, however, it is baffled by data fragmen-tation: logically consecutive chunks physically scattered across various containers. Many rewriting schemes, rewriting fragment-ed duplicate chunks into new containers, attempt to alleviate the restore performance degradation caused by fragmentation. Unfortunately, these schemes rely on a fixed threshold and fail to choose the appropriate set of old containers for rewriting, which leads to substantial redundant chunks existing in the retrieved containers when restoring backups. To address this issue, we propose a flexible threshold rewriting scheme to improve restore performance while maintaining high backup performance. We define an effectiveness metric - valid container reference counts (VCRC) - that facilitates identifying the appropriate containers for rewriting. We design a greedy-algorithm-based algorithm called F-greedy that dynamically adjusts the threshold according to the distribution of containers' VCRC, aiming to rewrite low-VCRC containers. We quantitatively evaluate F-greedy on three real-world backup datasets in terms of restore performance, backup performance, and storage overhead. The empirical results show that compared with two state-of-the-art schemes (Capping and SMR), our scheme improves the restore speed of the exiting algorithms by 1.3x - 2.4x while achieving similar backup performance.

Summary With the explosive growth of data, hundreds of thousands of servers may be contained in a single data center. Hence, node failures are unavoidable and generally negatively effects the performance of the whole data center. Additionally, data centers with a large number of nodes will cause plenty of energy consumption. Many existing task scheduling techniques can effectively reduce the power consumption in data centers by considering heat recirculation. However, the traditional techniques do not take the situation of node failures into account. This paper proposes an airflow‐based failure model for data centers by leveraging heat recirculation. In this model, the spatial distribution and time distribution of failures are considered. Furthermore, a genetic algorithm (GA) and a simulated annealing algorithm (SA) are implemented to evaluate the proposed failure model. Because the positions of node failures have a significant impact on the heat recirculation and the energy consumption of data centers, failures with different positions are analyzed and evaluated. The experimental results demonstrate that the energy consumption of data centers can be significantly reduced by using the GA and SA algorithms for task scheduling based on the proposed failure model.

Abstract In this study, we propose a caching method called RUE for dynamic large‐scale data streams. We define a data model to facilitate hot data identification and management. At the heart of RUE model is hot degree that takes into account two factors data resource utilization efficiency and reuse distance, aiming to quantitatively reflect data popularity in a dynamic data stream. Based on data's hot degree, RUE classifies data into four types, each of which is assigned with an associated cache residence time. Guided by RUE model, we develop HM algorithm to identify and manage hot data in a dynamic data stream. HM algorithm is implemented by four stacks, namely, new stack, short stack, long stack, and temp stack. Moreover, an eviction and a migration algorithms are integrated into HM to facilitate block replacement and migration. To evaluate the performance of HM algorithm, we quantitatively compare the performance of RUE with three state‐of‐art algorithms, namely, LRU, LIRS, and ARC under various replacement policies, operations, and workloads. Experimental results show that RUE outperforms these three existing algorithms in terms of both read and write hit rates. Furthermore, we show that with the four stacks in place, the computing overhead of HM is negligible.

To build energy-efficient datacenters, one widely used way is to dynamically manipulate the quantity of available hardware resources on demand. However, when bursty workloads appear, additional time overhead is required for resource gearing, thus incurring a performance degradation. To explore this problem, we present an intelligible analysis on the request handing in a VM. From the analysis, we find that, when workloads are overloaded (i.e., peak loads appear), the number of QoS guaranteed requests can be greatly increased by deferring the scheduling of a few requests. Inspired by this finding, we propose a Peak Load Scheduling Control (PLSC) method to promote the Quality of Service (QoS) of peak loads for modern energy-efficient datacenters. However, peak loads are usually difficult to identify. To overcome this difficulty, PLSC tracks the number of requests residing in a VM by leveraging a two-tier request queue maintained by it. When the number exceeds the capability of the VM, it means that peak loads appear. In this case, PLSC adds some delay-tolerant requests to the secondary queue. The scheduling of requests in the secondary queue is controlled with a lower priority than that of requests in the primary queue. Sequentially, with critical requests maintained in the primary queue, PLSC shortens the response time of critical requests. In addition, PLSC expands the number of QoS-guaranteed requests. Comprehensive experiments are conducted to attest the effectiveness of PLSC, by simulating a typical energy-efficient datacenter. The experimental results show that PLSC significantly promotes the QoS of workloads with a negligible impact on energy saving.

With the explosive data growth, an enormous number of computing and networking components (e.g., servers, switches, and wires) are continuously being augmented to data centers. Data center networks (DCNs) - maintaining a high network capacity - must be cost efficient, incrementally scalable, and fault-tolerant. To address these challenges, we propose in this study a new type of DCN architecture referred to asCriso. Different from the existing network architectures,Crisois designed hierarchically and recursively by employing two ports servers and commodity switches.Crisois constructed based on numerous isomorphicpods, each of which leverages external interfaces supplied by switches to connect with neighboring pods. Additionally, apod-based and fault-tolerant routing algorithm is designed to handle multiple failures.Crisohas an array of promising features, including being cost-efficient and delivering a high-network capacity that can be extended to millions of nodes. The analytic results demonstrate thatCrisois significantly superior to the four state-of-the-art data center structures in terms of network capacity, scalability, cost, power consumption, and other static characteristics. Furthermore, the experimental results unveil thatCrisosatisfies the fault-tolerant demands of modern data centers. Compared to the four existing topologies (i.e.,DCell,BCube,FiConn,Fat-Tree) that have been widely investigated,Crisois adroit at maintaining a balanced performance in terms of throughput and latency.

Data deduplication techniques construct an index consisting of fingerprint entries to identify and eliminate duplicated copies of repeating data. The bottleneck of disk-based index lookup and data fragmentation caused by eliminating duplicated chunks are two challenging issues in data deduplication. Deduplication-based backup systems generally employ containers storing contiguous chunks together with their fingerprints to preserve data locality for alleviating the two issues, which is still inadequate. To address these two issues, we propose a container utilization based hot fingerprint entry distilling strategy to improve the performance of deduplication-based backup systems. We divide the index into three parts: hot fingerprint entries, fragmented fingerprint entries, and useless fingerprint entries. A container with utilization smaller than a given threshold is called a sparse container . Fingerprint entries that point to non-sparse containers are hot fingerprint entries. For the remaining fingerprint entries, if a fingerprint entry matches any fingerprint of forthcoming backup chunks, it is classified as a fragmented fingerprint entry. Otherwise, it is classified as a useless fingerprint entry. We observe that hot fingerprint entries account for a small part of the index, whereas the remaining fingerprint entries account for the majority of the index. This intriguing observation inspires us to develop a hot fingerprint entry distilling approach named HID . HID segregates useless fingerprint entries from the index to improve memory utilization and bypass disk accesses. In addition, HID separates fragmented fingerprint entries to make a deduplication-based backup system directly rewrite fragmented chunks, thereby alleviating adverse fragmentation. Moreover, HID introduces a feature to treat fragmented chunks as unique chunks. This feature compensates for the shortcoming that a Bloom filter cannot directly identify certain duplicated chunks (i.e., the fragmented chunks). To take full advantage of the preceding feature, we propose an evolved HID strategy called EHID . EHID incorporates a Bloom filter, to which only hot fingerprints are mapped. In doing so, EHID exhibits two salient features: (i) EHID avoids disk accesses to identify unique chunks and the fragmented chunks; (ii) EHID slashes the false positive rate of the integrated Bloom filter. These salient features push EHID into the high-efficiency mode. Our experimental results show our approach reduces the average memory overhead of the index by 34.11% and 25.13% when using the Linux dataset and the FSL dataset, respectively. Furthermore, compared with the state-of-the-art method HAR, EHID boosts the average backup throughput by up to a factor of 2.25 with the Linux dataset, and EHID reduces the average disk I/O traffic by up to 66.21% when it comes to the FSL dataset. EHID also marginally improves the system's restore performance.

Large-scale clusters are often built with over-provisioned service resources, so as to satisfy the huge demand raised by enormous users in cloud environments. By estimating the resource demand of workloads, an on-demand resource provisioning method can be realized in these clusters, thus improving the energy efficiency. However, to guarantee Quality of Service (QoS), the resource demand of workload should be accurately estimated so as to provide suitable resources. Many statistical approaches estimate actual resource demand based on some workload features. But the relations between actual resource demand and workload features are generally obscure, and it's a big challenge to gain an accurate estimation under an obscure relation. In this paper, by considering a cluster as a queueing system, we construct a linearly dependent relation between resource demand and multiple feature combinations. The linearly dependent relation is inconstant due to its variable coefficients. Then, to ascertain specific relations which match actual situations, we design a Basic Linear regression (BL) algorithm. BL can obtain the optimal values for these coefficients, thus determining the inconstant relation to specific ones. Finally, we propose a Constructed Linear regression (CL) approach to estimate actual resource demands. CL forms a two-layer neural network by using several processes of BL as the neurons. To evaluate CL, we realize an On-Demand Resource Provisioning (ODRP) method in a typical power-aware cluster. Several evaluation metrics are proposed for conducting extensive experiments. The experimental results show that CL is effective to make accurate estimations.

Data centers consisted of a great number of IT devices (e.g., servers, switches and etc.) which generates a massive amount of heat emission. Due to the special arrangement of racks in the data center, heat-recirculation often occurs between nodes. It can cause a sharp rise in temperature of the equipment coupled with local hot spots in data centers. Existing VM placement strategies can minimize energy consumption of data centers by optimizing resource allocation in terms of multiple physical resources (e.g., memory, bandwidth, cpu and etc.). However, existing strategies ignore the role of heat-recirculation in the data center. To address this problem, in this study, we propose a heat-recirculation-aware VM placement strategy and design a Simulated Annealing Based Algorithm (SABA) to lower the energy consumption of data centers. Different from the existing SA algorithm, SABA optimize the distribution of the initial solution and the way of iteration. We quantitatively evaluate SABA’s performance in terms of algorithm efficiency, the activated servers and the energy saving against with XINT-GA algorithm (Thermal-aware task scheduling Strategy), FCFS (First-Come First-Served), and SA. Experimental results indicate that our heat-recirculation-aware VM placement strategy provides a powerful solution for improving energy efficiency of data centers.

As the volume of data keeps growing rapidly, more and more storage devices, servers and network devices are continuously added into data centers to store, manage and analyze the data. The industry experience indicates that, instead of a huge number of servers added at a time, the data center network also expands gradually by adding a small number of servers from time to time. As a result, how to achieve an incremental scalability is becoming a very important challenge in designing modern data center network architectures in order to maintain the topological properties unchanged when the size of data centers grows. In this paper, we propose a new type of data center network architecture named HSDC (High Scalability Data Center Network Architecture) based on the hypercube network. The HSDC is constructed by using $m$m-port switches and 2-port servers. The fault-tolerant routing algorithm designed in this paper for HSDC can be executed on any vertex and is able to construct a path between any pair of vertices. In order to achieve an incremental scalability, we further propose three types of incomplete HSDC structures that allow gradually adding servers into the structures, while maintaining all the topological properties. The simulation experiments and performance results demonstrate that the throughput of HSDC is comparable to that of Fat-Tree, BCube and DCell. Furthermore, the analysis results indicate that HSDC strikes a good balance among diameter, bisection width, incremental scalability, cost and energy consumption in contrast to the state-of-the-art data center network architectures.

With the data growing explosively, data center networks (DCN) have to possess the characteristics of incrementally scalable, cost-efficient, high network capacity and fault tolerance. However, the widely used DCNs can not meet the demands above. In this paper, we propose a new type of data center topology named Criso to settle the challenges. Different from the existed works, Criso has the advantages of both switch-centric topologies (servers do not participate in routing) and the server-centric topologies (the scalability is not limited by the ports of switches). It is constructed based on pods, The internal structure of each pod is the same and there are only four external interfaces. By applying such structure, a pod-based and fault-tolerant routing algorithm is designed to handle multiple types of failures. Criso is hierarchically, recursively defined and high-network capacity which can scale up to millions of nodes. The analysis results demonstrate that the Criso model is significantly superior to four state-of-the-art data center structures in terms of the network capacity, scalability, cost, power consumption and other static characteristics. Criso achieves the target of low-cost, low-energy consumption and highly-scalability simultaneously.

The explosive growth of Internet of Things is generating massive data which are normally stored in data centers. The power consumption has become a very important challenge of designing modern data centers due to the explosive growth of data. The power consumed by cooling system accounts for about half of the total power consumption. Reducing the peak inlet temperature of racks residing in data centers can effectively decrease the temperature requirement of supplied cold air, thus cutting down the cooling cost. Task distribution in data centers has a significant impact on this inlet temperature. Many investigations have been conducted on achieving an optimal task distribution in terms of the air organization [e.g., genetic algorithm (GA)]. However, the existing methods can be easily trapped into a local optimum. This paper constructs a power model to correlate the task assignment, heat recirculation, inlet temperature, and cooling cost in the homogeneous and heterogeneous data centers with under-floor air supply. Furthermore, genetic simulated annealing algorithm is proposed and designed to enhance the traditional GA and assign tasks in the data centers according to the corresponding air organization by integrating the advantages of simulated annealing, thus minimizing the inlet temperature and reducing the cooling cost. Experimental results indicate that the proposed approach can effectively decrease the temperature requirement of supplied cold air and reduce the power consumption of the cooling system in contrast to the traditional GA and ant colony algorithm, especially when the data centers are with medium utilization.

Big data has received considerable attentions in recent years because of massive data volumes in multifarious fields. Considering various “V” features, big data tasks are usually highly complex and computational intensive. These tasks are generally performed in parallel in data centers resulting in massive energy consumption and Green House Gases emissions. Therefore, efficient resource allocation considering the synergy of the performance and energy efficiency is one of the crucial challenges today. In this paper, we aim to achieve maximum energy efficiency by combining thermal-aware and dynamic voltage and frequency scaling (DVFS) techniques. This paper proposes: (a) a thermal-aware and power-aware hybrid energy consumption model synchronously considering the computing, cooling, and migration energy consumption; (b) a tensor-based task allocation and frequency assignment model for representing the relationship among different tasks, nodes, time slots, and frequencies; and (c) a big data Task Scheduling algorithm based on Thermal-aware and DVFS-enabled techniques (TSTD) to minimize the total energy consumption of data centers. The experimental results demonstrate that the proposed TSTD algorithm significantly outperforms the state-of-the-art energy efficient algorithms from total, computing, and cooling energy consumption perspectives, as well as cooling energy consumption proportion and total energy consumption savings.

The explosive growth of data brings new challenges to the data storage and management in cloud environment. These data usually have to be processed in a timely fashion in the cloud. Thus, any increased latency may cause a massive loss to the enterprises. Similarity detection plays a very important role in data management. Many typical algorithms such as Shingle, Simhash, Traits and Traditional Sampling Algorithm (TSA) are extensively used. The Shingle, Simhash and Traits algorithms read entire source file to calculate the corresponding similarity characteristic value, thus requiring lots of CPU cycles and memory space and incurring tremendous disk accesses. In addition, the overhead increases with the growth of data set volume and results in a long delay. Instead of reading entire file, TSA samples some data blocks to calculate the fingerprints as similarity characteristics value. The overhead of TSA is fixed and negligible. However, a slight modification of source files will trigger the bit positions of file content shifting. Therefore, a failure of similarity identification is inevitable due to the slight modifications. This paper proposes an Enhanced Position-Aware Sampling algorithm (EPAS) to identify file similarity for the cloud by modulo file length. EPAS concurrently samples data blocks from the head and the tail of the modulated file to avoid the position shift incurred by the modifications. Meanwhile, an improved metric is proposed to measure the similarity between different files and make the possible detection probability close to the actual probability. Furthermore, this paper describes a query algorithm to reduce the time overhead of similarity detection. Our experimental results demonstrate that the EPAS significantly outperforms the existing well known algorithms in terms of time overhead, CPU and memory occupation. Moreover, EPAS makes a more preferable tradeoff between precision and recall than that of other similarity detection algorithms. Therefore, it is an effective approach of similarity identification for the cloud.

The explosive growth in the volume of data storing and complexity of data processing drive data center networks (DCNs) to become incrementally scalable and cost-efficient while to maintain high network capacity and fault tolerance. To address these challenges, this paper proposes a new structure, called Totoro, which is defined recursively and hierarchically: dual-port servers and commodity switches are used to make Totoro affordable; a bunch of servers are connected to an intra-switch to form a basic partition; to construct a high-level structure, a half of the backup ports of servers in the low-level structures are connected by inter-switches in order to incrementally build a larger partition. Totoro is incrementally scalable since expanding the structure does not require any rewiring or routing alteration. We further design a distributed and fault-tolerant routing protocol to handle multiple types of failures. Experimental results demonstrate that Totoro is able to satisfy the demands of fault tolerance and high throughput. Furthermore, architecture analysis indicates that Totoro balances between performance and costs in terms of robustness, structural properties, bandwidth, economic costs and power consumption.

Over the past decade, many data centers have been constructed around the world due to the explosive growth of data volume and type. The cost and energy consumption have become the most important challenges of building those data centers. Data centers today use commodity computers and switches instead of high-end servers and interconnections for cost-effectiveness. In this paper, we propose a new type of interconnection networks called Exchanged Cube-Connected Cycles (ExCCC). The ExCCC network is an extension of Exchanged Hypercube (EH) network by replacing each node with a cycle. The EHnetwork is based on link removal from a Hypercube network, which makes the EHnetwork more cost-effective as it scales up. After analyzing the topological properties of ExCCC, we employ commodity switches to construct a new class of data center network models, namely ExCCC-DCN, by leveraging the advantages of the ExCCC architecture. The analysis and experimental results demonstrate that the proposed ExCCC-DCN models significantly outperform four state-of-the-art data center network models in terms of the total cost, power consumption, scalability, and other static characteristics. It achieves the goals of low cost, low energy consumption, high network throughput, and high scalability simultaneously.

Caching and prefetching are effective approaches to boosting the performance of metadata access in distributed storage systems. Many research efforts have been devoted in developing new metadata prefetching methods by considering past file access patterns. However, the existing methods do not consider the correlations between processes and the corresponding files(e.g. file provenance). Therefore, the methods cannot obtain very rich and accurate correlations, thus decreasing the effectiveness of metadata prefetching. This paper presents a Provenance-based Metadata Prefetching(ProMP) scheme, which considers both provenance and the past file access patterns. Through mining the correlations between processes and corresponding files from provenance and past access history, ProMP can achieve accurate and rich correlation information. ProMP is conducive to employing aggressive metadata prefetching to boost the performance by leveraging the correlations. Our experimental results show that ProMP performs more effectively with less memory overhead than the existing solutions, while improving the hit rates by up to 49% and 7% in contrast to traditional LRU and a state-of-art metadata prefetching algorithm Nexus, respectively.

Currently, novel topologies and advanced resource allocation strategies for multi-tenant cloud datacenters are two research hotspots. Due to the high convenience and efficiency, researchers tend to use simulation to evaluate the proposed topologies or strategies. However, the current network simulators do not support the multi-tenant cloud environment inherently. Moreover, the low simulation speed and high memory consumption limit the traditional packet-level simulators to estimate the scenario of large-scale datacenters. In this paper, we propose a new flow-level network simulator, mtCloudSim, to overcome the above issues. The simulator estimates the data flow's behavior in the real world, i.e., 1) increasing the sending rate when the network is not busy and 2) suspending when the congestion occurs. Bandwidth isolation is inherently provided and users are allowed to define bandwidth requirement for the experiments with our simulator. Object-oriented programming (OOP) makes it easy to evaluate novel network topologies. The tracing system is also able to generate abundant and detailed statistics for experiments. The experiments demonstrate that mtCloudSim is available for multi-tenant cloud evaluation.

In-line dedupliaction systems mainly focus on secondary storage for backup and archiving, and just offer several simple APIs. Applications can not directly invoke these APIs without modification. Although file systems offer abundant APIs and amity to applications, building a file system for in-line deduplication brings new challenges in the I/O path. Read operations involve multiple disk accessing, which includes getting fingerprints in file recipes, obtaining the addresses by checking fingerprint index and reading corresponding data blocks in disk drive. This extremely increases the latency in read path. We present a Low-Read-Latency File System (LRLFS) for the in-line deduplication. Experiments suggest that LRLFS obtains low read latency in read path with negligible storage overhead, acceptable CPU and memory utilization rate.

The Internet of Things has become a spotlight for a long period of time and generates massive amounts of sensor data. Thus, data centers play more and more crucial roles in processing and analyzing the explosively increasing data. To remedy the shortcomings of traditional tree-based structure, many novel server-centric network structures have been proposed in recent years. Their original routing mechanisms based on divide and conquer (DC) are not able to work out the shortest paths. So, there is still promotion room for communication delay reduction. Since dynamic programming (DP) is a classical strategy to obtain optimal solution, this paper proposes a routing mechanism based on DP and applies it to data center for better solving the weakness occurred by DC. Experiments firmly support the conclusion that adopting DP in routing calculation achieves appealing performance of short latency, great fault-tolerance and reasonable resource consumption. Theoretical analysis also proves that it is applicable to most popular structures.

Data deduplication has been widely used at data backup system due to the significantly reduced requirements of storage capacity and network bandwidth. However, the performance of data deduplication gradually decreases with the growth of deduplicated data. This is because the volume of fingerprints grows significantly with the increase of backup data, and a large portion of fingerprints have to be stored on disk drives. This incurs frequent disk accesses to locate fingerprints and blocks the process of data deduplication. Furthermore, the fingerprints belonging to the same file may be discretely stored on disk drives. This generates random and small disk accesses, and results in significant performance degradation when the fingerprints are referred. Additionally, a single fingerprint may appear only once during a backup process. This results in very low cache hit ratio due to lacking temporal locality. This paper proposes to employ file similarity to enhance the fingerprint prefetching, thus improving the cache hit ratio and the performance of data deduplication. Furthermore, the fingerprints are arranged sequently in terms of the backup data stream to maintain the locality and promote the performance. Experimental results demonstrate that the proposed idea can effectively reduce the number of fingerprint accesses going to disk drives, decrease the query overhead of fingerprints, thus significantly alleviating the disk bottleneck of data deduplication.

Variety in Big Data means we have a wide range of data types and sources: e.g. File systems and database systems co-exist for decades as two popular data-accessing interfaces. This work is to unify these two interfaces by presenting a Data Interface All-iN-A-place (DIANA). The first challenge lies in distinguishing structured and un-structured data and diverting them to different underlying platforms. It is demonstrated that a speedup of 5000 in indexing has been achieved at the expense of a slowdown of 100 in extracting attributes. A DIANA-based cloud storage system is constructed for versatile, long distance and large volume big data accessing operations to address "Volume" and "Velocity" in Big Data. It encapsulates a dynamic multi-stream/multi-path engine at the socket level, which conforms to Portable Operating System Interface (POSIX).

Many programs require more RAM to hold their data than a typical computer has. Theoretically, both the compression and deduplication can trade the rich computing capacity for more available RAM space. This paper comprehensively evaluates the performance behaviour of memory compression and memory deduplication by using seven real memory traces. The experimental results give two implications: (1) Memory deduplication greatly outperforms memory block compression. (2) Fixed-size partition (FSP) achieves the best performance in contrast to Content-defined Chunking (CDC) and Sliding Block (SB). The optimal chunking size of FSP is equal to the size of a memory page. The analysis results in this paper should be able to provide useful insights for designing or implementing systems that require abundant memory resources to enhance the system performance.

Energy efficiency has become one of the most important challenges in designing large-scale clusters due to the commercialized computer components. Traditionally, load balancers are employed by clusters to improve system performance and scalability. However, those balancers do not consider the energy used by the clusters. A power-aware cluster scheduler has been proposed in the community by concentrating the workloads to a few nodes, and turning off those nodes which do not have workloads. By augmenting the traditional cluster schedulers with power management, this paper explores the impacts of power management on the different cluster schedulers. Real traces are used to evaluate the impacts. Though the cluster schedulers are significantly different, from a power standpoint, we only observed slight difference of different cluster schedulers.

A grid-powered Web Geographical Information Science (GIS)/Web Processing Service (WPS) system has been developed for archiving and distributing large volumes of geospatial data. However, users, WPS servers, and data resources are always distributed across different locations, attempting to access and archive geospatial data from a GIS survey via conventional Hypertext Transport Protocol, Network File System Protocol, and File Transfer Protocol, which often encounters long waits and frustration in wide area network (WAN) environments. To provide a “local-like” performance, a WAN/grid-optimized protocol known as “GridJet” developed at our lab was used as the underlying engine between WPS servers and clients, which utilizes a wide range of technologies including the one of paralleling the remote file access. No change in the way of using software is required since the multistreamed GridJet protocol remains fully compatible with the existing IP infrastructures. Our recent progress includes a real-world test that PyWPS and Google Earth over the GridJet protocol beat those over the classic ones by a factor of two to eight, where the distribution/archiving distance is over 10 000 km.

In the above-named work the name of one of the authors is incorrectly given. Included here also is the biography with the missing IEEE membership information.

This paper describes the grid-oriented storage (GOS) architecture and its implementations. A GOS-specific file system (GOS-FS), the single-purpose intent of a GOS OS, and secure interfaces via grid security infrastructure (GSI) motivate and enable this new architecture. As an FTP server, GOS with a slimmed OS, with a total volume of around 150 MB, outperforms the standard GridFTP by 20-40 percent. As a file server, GOS-FS acts as a network/grid interface, enabling a user to perform searches and access resources without downloading them locally. In the real-world tests between Cambridge and Beijing, where the transfer distance is 10,000 km, the multistreamed GOS-FS file opening/saving resulted in a remarkable performance increase of about 2-25 times, compared to the single-streamed network file system (NFSv4). GOS is expected to be a variant of or successor to the well-used network-attached storage (NAS) and/or storage area network (SAN) products in the grid era