Multimodal neural networks (MMNNs) feature multi-branch topologies that offer opportunities for inference parallelism. However, mainstream machine learning compilers prioritize intra-operator parallelism as the primary optimization target. While effective for chain-structured models, this approach inhibits the inter-operator parallelism inherent in the MMNNs, leading to suboptimal inference efficiency. Through a systematic analysis, we identify two fundamental sources of inefficiency: (1) during graph scheduling, a compromise among synchronization overhead, resource contention, and GPU utilization; and (2) during compilation, a structural conflict between greedily maximizing intra-operator parallelism and the severe resource competition induced by concurrent operators.

Aortic Dissection (AD) is a life-threatening disease that can be rapidly screened by using deep learning methods. However, deep learning model training requires a large amount of manual annotation of data. To improve the annotation efficiency and accuracy, we propose LABEL-SAM, a semi-automatic interactive segmentation algorithm designed to efficiently annotate AD in 3D computed tomography angiography (CTA) images. By requiring minimal user input—only points and bounding boxes on the first and last slices—LABEL-SAM automatically generates segmentation prompts for intermediate slices, reducing the manual workload. In addition, we propose a bidirectional prediction weighting method and a fine-tuning strategy tailored to AD data, further enhancing segmentation accuracy. Furthermore, LABEL-SAM is implemented as a plug-and-play plugin for the 3D Slicer software. Experimental results on both external and internal datasets demonstrate the method’s superior performance, improving annotation accuracy and efficiency. The code will be available at https://github.com/wenjiecai/LABEL-SAM. The demonstration video is now available at https://www.youtube.com/watch?v=R3Fzgl1b4JQ.

Aortic Dissection (AD) is a life-threatening disease. As such, it’s vital to make clinical strategies by accurate segmentation of AD from Computed Tomography Angiography (CTA) images. Due to the heterogeneous morphology of AD in CTA images and the influence of image noise, inaccurate segmentation results are often produced, making it challenging to accurately identify the contours of AD. To meet these challenges, we propose a novel automatic AD segmentation model DETR-SAM based on the Segment Anything Model(SAM). Our model consists of a deformable DETR object detector module with multi-scale feature fusion and a fine-tuned SAM module for refining the segmentation contours. The object detector module effectively fuses the multi-scale image features extracted by the image encoder of SAM with the features obtained from the DETR backbone CNN network. The fused multi-scale features generate multiple prompts rich in geometric shape knowledge through multiple regression prediction heads to guide the SAM model to refine the segmentation boundaries and enhance the recognition of contours. Meanwhile, a fine-tuning strategy using these prompts is proposed to enhance the SAM model’s understanding of AD features. This strategy enables our DETR-SAM model with a 2D segmentation network architecture to surpass the performance of mainstream 3D segmentation algorithms. We have evaluated DETR-SAM on the ImageTBAD dataset and our internal dataset. The results showed that the accuracy of our model is better than the current state-of-the-art 2D and 3D segmentation methods in all four indicators. In addition, the effectiveness of the feature fusion module, hint generation method and fine-tuning strategy are further verified by ablation experiments. Our DETR-SAM method is helpful to a certain extent in the diagnosis and auxiliary treatment of AD in different clinical situations and lays a foundation for further research and clinical application.

Due to the susceptibility to nothing-at-stake and long-range attacks, the Proof-of-Stake consensus faces challenges in securely and efficiently confirming blocks. We propose a new Proof-of-Stake consensus, Voted Weightest Sub-Tree(VWeiST) consensus. It assigns weights to each block through voting, and nodes confirm blocks by calculating the probability that each block's weight can be exceeded by other competitors. We employ a multi-round voting approach, where a small number of nodes are randomly selected as the committee nodes to vote in each round. This approach results in particularly low communication overhead per block, allowing for scalability to a large number of nodes. Compared to other consensus, our mechanism requires fewer rounds of voting to confirm a block, offering advantages in throughput and transaction latency. In the experiments, VWeiST achieves latency and round reductions down to 40% and 29% of the comparison method's levels at most. Furthermore, we theoretically prove that the consensus ensures liveness and probabilistic safety.

Traditional blockchain systems still struggle with limited throughput, particularly those compatible with EVM, which are crucial in many blockchain applications. One of the main reasons arises from serial execution, which doesn’t exploit parallelism in transaction execution. Although some recent literature introduced concurrency control mechanisms to execute transactions in parallel, they do not work efficiently in real-world blockchains where proposers and validators have different execution contexts, which means varying execution deterministic levels and execution quantities.

Blockchain, a widely utilized distributed ledger technology, faces the scalability challenge. State sharding has emerged as a promising solution for addressing this challenge. However, conventional state allocation solutions often face two major obstacles: a high ratio of cross-shard transactions and an unbalanced workload distribution, due to their reliance on a simple and fixed assignment of states to shards. The former increases the overall workload on the system, while the latter reduces resource utilization. Both factors significantly impact system performance. Moreover, our key observation is that the collection of smart contract transactions can be represented as a hypergraph network by analyzing their characteristics. Therefore, this study proposes HyperChain, a novel dynamic state sharding protocol that integrates a hypergraph partition algorithm. HyperChain aims to reduce the ratio of cross-shard transactions and balance workload distribution, thereby achieving improved throughput and reduced transaction latency in smart contract blockchain systems. Our experiments demonstrate that the proposed HyperChain exhibits superior performance than other solutions in terms of cross-shard transaction ratio, workload balance, throughput, and transaction latency.

Satisfiability Modulo Theories (SMT) solvers have been widely applied in automated software analysis to reason about the queries that encode the essence of program semantics, relieving the heavy burden of manual analysis. Many SMT solving techniques rely on solving Boolean satisfiability problem (SAT), which is an NP-complete problem, so they use heuristic search strategies to seek possible solutions, especially when no known theorem can efficiently reduce the problem. An emerging challenge, named Mixed-Bitwise-Arithmetic (MBA) obfuscation, impedes SMT solving by constructing identity equations with both bitwise operations (and, or, negate) and arithmetic computation (add, minus, multiply). Common math theorems for bitwise or arithmetic computation are inapplicable to simplifying MBA equations, leading to performance bottlenecks in SMT solving.

In sub-second stream computing, the answer to a complex query usually depends on operations of aggregation or join on streams, especially multi-way theta join. Some attribute keys are not distributed uniformly, which is called the data intrinsic skew problem, such as taxi car plate in GPS trajectories and transaction records, or stock code in stock quotes and investment portfolios etc. In this paper, we define the concept of key redundancy for single stream as the degree of data intrinsic skew, and joint key redundancy for multi-way streams. We present an execution model for multi-way stream theta joins with a fine-grained cost model to evaluate its performance. We propose a solution named Group Join (GroJoin) to make use of key redundancy during transmission and execution in a cluster. GroJoin is adaptive to data intrinsic skew in the way that it depends on the grouping condition we find out, i.e., the selectivity of theta join results should be smaller than 25%. Experiments are carried out by our MS-Generator to produce multi-way streams, and the simulation results show that GroJoin can decrease at most 45% transmission overheads with different key redundancies and value-key proportionality coefficients, and reduce at most 70% query delay with different key distributions. We further implement GroJoin in Multi-Way Stream Theta Join by Spark Streaming. The experimental results demonstrate that there are about 40%~50% join latency reduced after our optimization with a very small computation cost.

In large-scale Internet of Thing (IoT), ambient backscatter communication has become a new green technology of concern. At present, the research scenario on ambient backscatter communication mainly focuses on single-transmitter and single-receiver, and a few studies have multiple receivers. And backscatter system is always only allowing one transmitter-receiver pair active. However, the condition multiple transmitters communicate with multiple receivers is essential to achieve giant connection in Internet of Everything. Besides, the system with active multipair can transmit more byte and use energy more effectively. So there is an urgent need of research on multi-transmitter multi-receiver system. We propose an ambient back scatter communication system which allows multi-transmitter and multi-receiver active. This paper is devoted to studying the user association problem in such system. When studying the ambient backscatter communication system, we pay attention to the power limitation. Because the energy collected by the device is relatively small and limit communication performance. In the case of limited link budget, this paper maximized system communication capacity. A priority-based access strategy is proposed in this paper. It arranges priority to the receiver according to the power threshold and link budgets. Then, the strategy handles association problem according to receiver's priority from high to low. Simulation results show that the proposed access strategy has better convergence than the random access strategy. It achieves maximum communication capacity and suboptimal bit rate. What's more, it has low complexity.

Acceleration in deep reinforcement learning has attracted the attention of researchers. Many parallel training frameworks have been proposed to speed up the sampling and the training by running multiple agents simultaneously. However, the bottleneck of performance in a single agent is the neural network prediction which takes a lot of time as compared to the rapid environmental changes. Different from these parallel frameworks, we try to speed up the prediction of the agent to accelerate the entire training process. As far as we know, this is the first time to accelerate the training from this perspective. We propose a novel training framework NNC-DRL to accelerate the whole training process of deep reinforcement learning. NNC-DRL uses different neural networks to represent target policy and behavior policy. The behavior policy network is a smaller neural network derived from the target policy network, which could speed up the prediction. The target policy network will transfer its latest policy to the behavior policy network. The inconsistent policy distribution between behavior network and target network will degrade the convergence of the training. In NNC-DRL, we introduce the Important Sampling technique to estimate policy gradient, which could improve the convergence of the training. The experiments show that our approach NNC-DRL can speed up the whole training process by about 10-20% on Atari 2600 games with little performance loss.

In cloud datacenter, it should be rational to enforce fair allocation on network resources among VDCs (virtual datacenters) in terms of multi-tenant model. Traditionally, cloud networks are shared in a best-effort manner, making it hard to reason about how network resources are allocated. Prior works concentrate on either providing minimum bandwidth guarantee or achieving work-conserving based on the VM-to-VM flow policy or per-source policy, or both. However, fair allocation on redundant bandwidth among VDCs is ignored. In this paper, we design NXT-Freedom, a bandwidth guarantees enforcement framework that divides network capacity based on per-VDC fairness while achieving work-conservation. To ensure per-VDC fair allocation, a hierarchical max-min fairness algorithm is proposed. To be applicable to non-congestion-free network core and to be scalable, NXT-Freedom decouples computing per-VDC allocation from enforcing the allocation. Through evaluation of a prototype, we show that NXT-Freedom achieves per-VDC performance isolation, and can be rapidly adapted to flow variation in cloud datacenter.

Interactive mobile applications attract lots of attentions recently. They utilize complex algorithms (e.g., machine learning) to provide advanced functions (e.g., object recognition), thus lead to long response time while running on mobile devices. To reduce the response time, researchers propose offloading some compute-intensive parts of mobile applications onto cloud. Existing works aim to optimize general performance (e.g., response time), but ignore the enhancement of application quality (e.g., recognition accuracy), which is also critical to user experience. In this paper, we develop AppBooster, a mobile cloud platform which boosts both general performance and application quality for interactive mobile applications. AppBooster jointly leverages the quality adaptation, computation offloading and parallel speedup to boost the comprehensive performance, which is defined by developers based on the metrics of application quality and general performance. Through combining history-based platform-learned knowledge, developer-provided information and the platform-monitored environment conditions (e.g., workload, network), AppBooster manages applications with optimal computation partitioning scheme and tunable parameter setting thus obtain high comprehensive performance. We evaluate AppBooster with an object recognition application in various network conditions and show AppBooster can significantly boost application performance and obtain 1.3 to 3.5 times better performance than existing strategies.

Improving performance of a mobile application by offloading its computation onto a cloudlet has become a prevalent paradigm. Among mobile applications, the category of interactive data-streaming applications is emerging while having not yet received sufficient attention. During computation offloading, the performance of this category of applications (including response time and throughput) depends on network latency and bandwidth between the mobile device and the cloudlet. Although a single cloudlet can provide satisfactory network latency, the bandwidth is always the bottleneck of the throughput. To address this issue, we propose to use multiple cloudlets for computation offloading so as to alleviate the bandwidth bottleneck. In addition, we propose to use multiple module instances to complete a module, enabling more fine-grained computation partitioning, since data processing in many modules of data-streaming applications could be highly parallelized. Specifically, at first we apply a fine-grained data-flow model to characterize mobile interactive data-streaming applications. Then we build a unified optimization framework that achieves maximization of the overall utilities of all mobile users, and design an efficient heuristic for the optimization problem, which is able to make trade-off between throughput and energy consumption at each mobile device. At the end we verify our algorithm with extensive simulation. The results show that the overall utility achieved by our heuristic is close to the precise optimum, and our multiple-cloudlet mechanism significantly outperforms the single-cloudlet mechanism.

Resource prediction (e.g. CPU/memory utilization) of cloud computing jobs has attracted substantial amount of attention. Existing works use regression methods based on historical information of jobs, with an impractical assumption that the job to be predicted has the same class as the historical jobs. To address this problem, we propose to take the category of the jobs into consideration for effective resource prediction. Existing works on job classification either ignores the temporal variance of resource consumption during job execution or use it in a naive way, resulting in unsatisfactory classification accuracy and/or slow speed. In this paper, we introduce a new and efficient job classification approach, called Bejo. Inspired by the textual document classification methods, which use distribution of text words to describe and classify a document, Bejo treats the job as a document, assigns each collected resource consumption snapshot to a certain "resource word", and uses the distribution of the words to describe and classify a job. An ℓ1norm minimization formulation is used to assign each resource snapshot to a resource word, to especially address the unique challenges of high noise and tight time budget of cloud job classification. We collect a comprehensive dataset for job classification and resource consumption prediction on cloud platforms, and demonstrate superior quality and efficiency of Bejo over state-of-the-art algorithms. Experiments also show the relative error of resource consumption prediction can be dramatically reduced by adding an extra job classification step to the existing regression methods.

In this paper, we propose a novel concept called "ubiquitous searching", which allows people to organize and search the desired information about the objects in the physical world, navigating from one object to others through their contextual links, just like what we do in the web searching. How to realize such an exciting idea poses many challenges, and a new approach is needed to provide scalable system abstractions and infrastructures. We identify the design principles and propose the framework toward ubiquitous search. The system architecture and the key algorithms for the proposed framework are developed. We also describe a proof-of-concept prototype and a simulation study used to experiment with our framework and algorithms.

Mobility management is an important issue for publish/subscribe systems to support mobile clients. The objectives of mobility management for publish / subscribe are to achieve short handoff delay and low message overhead, while at the same time guaranteeing reliable message delivery. Although mobility management has been extensively studied, the indirect communication style of publish/subscribe systems brings new challenges in designing mobility management solutions. In this paper, we propose a reliable and high-performance mobility management protocol, called multi-hop handoff (MHH) protocol, which sufficiently meets the requirements of publish/subscribe systems. A prototype was implemented for the MHH protocol and experiments were performed to compare the performance of MHH with two representative existing protocols. The experimental results demonstrate the efficiency improvement made by the proposed MHH protocol over the existing protocols.

Publish/subscribe (pub/sub) systems are very suitable for the dissemination of dynamic information over the Internet. As dynamic information is usually characterized by a short lifetime, both publishers and subscribers may specify the delay requirement on message delivery. Although existing pub/sub systems can easily be extended so that publishers and subscribers can specify their delay requirements, it remains a challenging problem to improve the efficiency of pub/sub systems so that as many messages can be successfully delivered as possible, while the network traffic does not increase significantly. In this paper, we propose an efficient approach for pub/sub systems to achieve bounded delay on message delivery. Three message scheduling strategies are proposed for the system to make use of available bandwidth efficiently. Simulation results show that our strategies enable subscribers to receive significantly more valid messages than traditional strategies, while the network traffic just increases slightly

In this paper, we propose a reliable multicast protocol for mobile agent communications, which is a multicast extension to our previous adaptive and reliable protocol (ARP). ARP uses the mailbox-based scheme, where each mobile agent is associated with a mailbox for message delivery. The mailbox can be decoupled from the mobile agent and migrate with lower mobility. By adjusting the system parameters such as synchronization between mobile entities and mailbox migration frequency, ARP achieves adaptability and ensures reliability. In extending ARP to group communications, we group mailboxes residing on a host and belonging to the same multicast group into a cluster, and let all mailbox clusters form a logical ring. We present a token-based algorithm and a mailbox migration algorithm to achieve atomicity and total ordering of multicast messages. Also, we prove the correctness of the protocol and discuss the performance evaluation results obtained from simulations.

The publish/subscribe paradigm provides a loosely coupled form of interaction that is well suitable for the large-scale distributed systems. In this paper, we introduce the semantic Web technologies into the publish/subscribe systems and propose a new type of publish/subscribe system. In the system, concept models are represented as ontologies, events are represented as RDF graphs, and subscriptions are represented as RDF graph patterns. The system can match events with subscriptions based on the semantics of events, and support events with complex data structures such as graphs

Based on business processes, composite Web services combine the offerings of two or more Web services to achieve the desired business goals. Several candidate standards have been proposed, providing a foundation for composite Web service specifications. However, at a higher level, there is no framework that supports composite Web service construction and automation. We propose a framework that facilitates the visual design, validation and automation of composite Web services. The framework is based mainly on Web service composition graph (WSCG), the underlying formalism for composite Web services. Using graph grammar and graph transformation defined on WSCG, the static topological structure of a composite Web service can be described in an intuitive way and the automation of the constructed composite Web services is also facilitated with a sound formal semantic basis. We also outline the design and implementation of a prototype for the proposed framework.

Recent advances in wireless and mobile networks have led to the exponential growth of mobile applications. Unlike conventional computing, mobile computing has stringent constraints in network resources, such as bandwidth and connectivity. As such, data in mobile applications are often cached at clients to increase performance, data availability and reliability. Formal verification of cache coherence in data access is essential in ascertaining the validity of a cache coherence protocol. Although a number of studies have been made in this subject, few researchers focused on mobile data access. In this paper, we present an automatic approach towards formal validation of a cache validation protocol supporting mobile data access. This approach combines the flexibility of visual modeling techniques with the rigor of formal validation. As it is difficult to construct the formal model of protocol, we have developed a set of formalization and translation rules to automate the process of construction. The reliability of the protocol has been verified using model checking.