Technologies such as intelligent reflective surface (IRS)-assisted free-space optics (FSO) systems, are capable of providing connectivity with varied bandwidth through flexibly allocating resources on a planer panel. This flexibility provides new opportunities in handling inter-rack traffic that is often unevenly distributed in space and time. In this paper, we explore a new type of resource allocation scheme by taking full advantage of this flexibility. The key ideas include: (1) dynamically divide the resource on the planar panel into two planes, such that the small and large flows can be handled separately; (2) in the plane serving small flows, use multi-hop routing with variable bandwidth connections to handle skewed traffic between source-destination pairs; (3) in the plane serving large flows, use direct links with variable bandwidth to schedule unevenly distributed demand. In essence, through flexible resource allocation in and between the two planes, this scheme is able to make more efficient use of resources. We theoretically analyze the capacities of the proposed scheme, where the effectiveness conditions and performance bounds for dynamic resource adaptation are derived and the traffic scheduling with flexible bandwidth in two planes is proved to be able to achieve constant factor approximation to the ideally optimal performance. We then propose a set of lightweight algorithms with performance guarantees in each of two planes, collectively calledFlexAlloc, to realize the flexibility and adaptivity of the proposed scheme, and discuss an example of this scheme in the context of IRS-assisted FSO systems. Our simulation results show thatFlexAllocexhibits significant superiority over existing schemes for unevenly distributed traffic at low computational complexity, and can reduce the average flow completion time by 56-81%.

When optical switching is deployed in Data Center Networks (DCNs), the reconfiguration of the optical switching matrix leads to substantially longer overheads, posing a significant impact on the system performance. Despite the extensive studies on the scheduling algorithms based on demand matrix decomposition (DMD), the stateful and irregular nature of the scheduling processes hinders the development of quantitative models, thereby limiting our understanding of resource scheduling in optical switching DCNs based on DMD. In this paper, we model the DMD based resource scheduling process under a bursty and skewed traffic pattern and derive closed-form equations for the burst completion time. Our study shows that an increased reconfiguration delay will lead to an approximate linear increase in the burst completion time. Our study also demonstrates that the size of the slot and the maximum allowed duration of one match are approximately inversely proportional to the burst completion time, with diminishing marginal returns.

The growing threat of physical inactivity to human health has highlighted the rising importance of community sports organizations, which play a crucial role in promoting physical activity and improving overall public health. However, the limited understanding of human behavior leads to suboptimal resource utilization, highlighting the necessity for a scientific understanding of human behavior within such social systems to optimize resource allocation, enhance participation levels, and improve public health. In this study, we aim to gain insights into the possible mechanisms behind user participation in community sports activities and reveal the role of external incentives on individual engagement. We study the participation behavior of 2,956 members over a 6-year period in a community sports organization, and demonstrate that the participation frequency adheres to a power-law distribution. We then propose a mathematical model called habit formation and behavioral inertia (HFBI), and through experiments, we demonstrate that the data generated by the model shows a robust fit to the distribution of empirical data. We also analyze the effects of different types of activities, and find that periods of high activity participation often begin with rewarded activities, suggesting the importance of incentives to sustain long-term participation. Our study provides a scientific foundation for developing optimized strategies to enhance community sports participation. The HFBI model may also be used to understand data exhibiting power-law distributions beyond the domain of community sports.

Entropy modeling is the core component of learned image compression (LIC) that models the distribution of latent representation learned from input images via neural networks for bit-rate estimation. However, existing entropy models employ presumed parameterized distributions such as Gaussian models and are limited for the learned latent representation characterized by complex distributions. To address this problem, in this paper, we for the first time achieve generative probabilistic entropy modeling of latent representation based on conditional diffusion models. Specifically, we propose a conditional diffusion-based probabilistic entropy model (CDPEM) to parameterize the latent representation with distributions of arbitrary forms that are generated by well designed training-test consistent denoising diffusion implicit model (TC-DDIM) without introducing any presumption. TC-DDIM is designed to leverage ancestral sampling to gradually approximate the distribution of latent representation with guaranteed consistency in generation for training and test. Furthermore, we develop a hierarchical spatial-channel context model to incorporate with TC-DDIM to sufficiently exploit spatial correlations with the approximate contextual information produced by ancestral sampling and channel-wise correlations using channel-wise information aggregation with reweighted training loss. Experimental results demonstrate that the proposed entropy model achieves state-of-the-art performance on the Kodak, CLIC, and Tecnick datasets compared to existing LIC methods. Remarkably, when incorporated with recent baselines, the proposed model outperforms latest VVC standard by an evident gain in R-D performance.

Intelligent reflective surface (IRS)-assisted free-space optics (FSO) systems allow allocation of different number of IRS elements between communication end-points. It thus can provide connectivity with variable bandwidth in data centers, where the traffic between top-of-rack switches is often unevenly distributed in space and time. In this paper, we propose a novel traffic scheduling technique that allows variable bandwidth in the context of IRS-assisted FSO systems in data center networks (DCNs). We formalize the problem of scheduling with variable bandwidth, explore the performance space for designing scheduling algorithms, and provide insights on how to use heuristics in practice. Then, we propose a low-complexity and performance-guaranteed parallel scheduling algorithm,pVarband, that takes advantage of variable bandwidth to avoid the time-consuming maximum weight matching and can run at a complexity of$\mathcal {O}(N\log N)$in a DCN withNracks. We theoretically demonstrate that scheduling with variable bandwidth requires at most$\mathcal {O}(\log (1/\epsilon))$reconfigurations to obtain an$\epsilon $-approximate decomposition for an arbitrary traffic matrix. Also, we provide insights for design and deployment of an IRS-assisted FSO system to achieve performance saturation. Our results show that scheduling allowing variable bandwidth can achieve almost optimal performance for unevenly distributed traffic, and also outperforms its fixed bandwidth counterpart in terms of running time and decomposition error. Our work also provides useful insights for scheduling with variable bandwidth in contexts other than IRS-assisted FSO systems.

Steerable free-space optics (FSO) communications are flexible in link reconfiguration (LR), and easy to deploy, especially when the available physical space is limited. Thus it is considered as a good complement to the wired network in data centers. In this paper, we are interested in understanding the role of FSO when used as a hotspot offloading technique in data center networks. We aim to quantify the performance of FSO-based hotspot offloading, in the presence of link impairment, failed negotiations and reconfiguration overhead. We model this steerable FSO-based traffic offloading process as a vacation queueing system with limited service discipline. A general expression for the average delay is derived, and then equations for three different hybrid automatic repeat request (HARQ) protocols are separately derived. To overcome the inherent difficulty of calculating the average delay with limited service, efficient low-complexity approximations are developed. We then show that the system delay increases approximately linearly with the traffic rate until a point, beyond which it grows rapidly. Further, we analyze the effect of various factors on the delay performance and offer insights for the design configurations to achieve the desired performance. The simulation results verify the validity of the theoretical analysis.

Deep learning models have greatly improved the accuracy of point cloud classification. Nevertheless, existing deep learning models are vulnerable to data corruptions such as Gaussian noise and outliers which are inevitable in real-world point cloud collection. To address this challenge, we develop a novel point cloud classification model that is robust to data corruptions, with permutohedral lattice-based representation and density-improved hierarchical feature extraction. Specifically, raw noisy point clouds are firstly projected into a regular per-mutohedral lattice space to obtain the quantized representations. Subsequently, we propose to leverage the local density to improve the farthest point sampling (FPS) and spectral graph convolution for robust hierarchical feature extraction, inspired by that the local point density implicitly reveals the reliability and semantic information of each point. The local density can be efficiently calculated from the permutohedral lattice-based representation. Extensive experiments on the benchmark dataset (i.e., ModelNet-C) verify the robustness of the proposed model. In comparison to the state-of-the-art methods, the proposed model achieves superior performance on the corrupted dataset while maintaining competitive classification accuracy on the clean dataset.

Entropy modeling plays an important role in estimating the rates of latent representations and optimizing the rate-distortion performance for learned image compression. Autoregression modules are demonstrated to eliminate spatial/channel-wise redundancy of latent representations in fixed-rate learned image compression. However, it cannot be efficiently achieved in progressive coding due to the high computational complexity raised by element-wise probability prediction. In this paper, we propose a learned progressive image compression method that enables spatial autoregression for entropy modeling. Specifically, we develop a novel codeword alignment scheme to prevent coding redundancy and achieve efficient autoregression of latent representations in different quality layers. Consequently, conditional probability estimation for the latent prediction can be achieved based on spatial autoregression in a layer-wise manner. We further extend the proposed method by dead-zone quantizers to obtain promoted rate-distortion performance. The proposed method is a successful attempt to enable spatial autoregression in learned progressive coding and further bridge the performance gap with fixed-rate models. Experimental results show that it outperforms traditional methods such as JPEG and BPG, as well as recent fine-grained learned progressive coding models DPICT and PLONQ in terms of rate-distortion performance.

As an accurate measurement of physical activity, step counts data can be collected expediently by smartphones and wearable devices. Complete and high time-resolution step counts data record the time and intensity of individuals’ physical activity in a day, and can be used to mine activity habits or to recommend customized workout plans. However, sparse step counts data are common in practice due to hardware and software limitations. Understanding the value of sparse step counts data can contribute to its application in healthcare, and also can help us design cost-effective hardware and software. In this article, we aim to infer activity patterns from sparse step counts data. We design a deep learning model based on recurrent neural networks, namely MLP-GRU, which considers bidirectional short-term dependency and long-term regularity of sparse step counts data, and implements data-driven imputation and classification. We also develop an interpretable and elastic method to obtain sparse step counts data labeled with multi-granular activity patterns to train MLP-GRU. Evaluations on real-world datasets reveal that MLP-GRU outperforms other strong baseline methods. The results also show that activity patterns can be inferred from extremely sparse step counts data with high accuracy, provided that proper granularity is used for data of different sparsity.

The data center markets in emerging economies are being built at a furious pace. When high availability is required, as it always is in the modern digital economy, the placement of geo-distributed data centers may be influenced by factors such as technician shortage and under-developed infrastructure, both of which are typical in emerging economies. Although the data center availability subject in general has been well studied, it remains unclear how rapid and unbalanced economic development in emerging economies may affect the availability of geo-distributed data centers and their cost of ownership. In this paper, we incorporate the unbalanced availability of infrastructure and technician into the data center placement. The problem is first formulated as a mixed integer nonlinear program (MINLP).To solve this potentially large scale problem, we transform it into a QCQP, capable of handling heterogeneous workloads. The resulting problem can then be efficiently solved by off-the-shelf optimization toolboxes. With real-life data in China, we show how unbalanced development of infrastructure and technician shortage may affect the placement of data centers, and analyze the tradeoff between cost and availability. Our results indicate that technician shortage and unbalanced network infrastructure will lead to increased cost and distinct data center placement strategies.

The increasing demand of bulk data transfer imposes significant challenges in inter-datacenter networks that span multiple domains. Storing data temporarily at domain borders and forward it at later off-peak hours (SnF) can reduce congestion and improve network utilization. However, performance gains from SnF transfer still requires further quantitative study. In this paper, we model deadline-constrained SnF transfers in networks with congested inter-domain links, where traffic has slightly varying peak hour and diurnal patterns. A closed-form estimation for the two-links case is derived to quantify the performance gain of intermediate storage (IS) and edge storage (ES). Our study shows that if bulk data transfer requests arrive with equal probability in a day, increase the allowed waiting time will increase the successful transfer probability linearly. Increasing the aggregated network load, defined as the summed duration of the background traffic peak-hour and data transfer, will result in linear decrease in successful transfer probability, when the network is moderately loaded. Our research also shows that IS has significant advantages over ES under heavy network load, and the maximum benefit occurs when the allowed waiting time is two times the aggregated network load.

As people become more health-conscious, more community sports organizations (CSOs) are emerging in society. The development of CSOs is often subject to very limited resources, and understanding its underlining principles is critical to maximizing the utility of available resources. Existing studies of CSOs are often empirical and adopt top-down methods of organizational management. This is partly because the inputs needed for such studies, for example, the interactions between community members, are often difficult to record. In this article, we perform an objective social network analysis on a real CSO with 2073 active members, with data collected over a five-year span on a mobile platform. Our study shows that the CSO members’ network has a low density and weak densification and is assortative and obvious in community structure. With a classification algorithm called SC-CEE, we observe that the development of the CSO can be robustly divided into stages. We further observe that the relationship between specific stimulus and community evolution varies across development stages, implying that an organization’s response strategy to change should be aligned with its stage of development. We then give practical guidance on the developmental priorities and response strategy of CSOs at different stages.

Emerging online services have fueled unprecedented demands for multicast transfers across geo-distributed datacenterse However, the time- and space-varying nature of background traffic in inter-datacenter networks (inter-DCNs) makes conventional multicast trees difficult to fully utilize the residual bandwidth in the inter-DCNs. In this paper, we introduce datacenter storage into multicast transfers, and propose Store-and-Forward-Tree (SnFTree) scheduling method for multicast transfers in the software-defined optical network. Compared to the conventional multicast scheduling methods, the advantages of SnFTree are as follows: (i) by enabling temporary storage at intermediate sites, SnFTree can segment the conventional multicast tree into independent subtrees, which maximizes the flexibility of provisioning; (ii) by introducing load-balancing routing into the tree computation, SnFTree can balance the traffic load across network efficiently; (iii) benefiting from optical multicast, storing, forwarding and reception can be simultaneously implemented on a site with negligible processing delay; (iv) by formulating the multicast scheduling problem into a routing problem, SnFTree greatly simplifies the problem for dynamic traffic. Simulations demonstrate that compared to conventional scheduling methods, SnFTree obtains lower blocking probability and requires fewer OpenFlow Agent Updates.

Traffic fluctuations make conventional multicast tree difficult to fully utilize the residual bandwidth in the inter-datacenter networks. In this paper, we introduce datacenter storage into the provisioning process of multicast transfers. A scheduling method, named Store-and-Forward-Tree (SnFTree), is proposed for multicast transfers in the software-defined inter-datacenter optical network. By leveraging temporary storage at intermediate sites, a multicast tree can be segmented into time-independent subtrees when provisioning an entire light-tree fails. Benefiting from the broadcast nature of optical multicast, intermediate sites can provide cut-through routing and temporary storage simultaneously with negligible processing delay. By using load-balancing tree computation, traffic can be balanced across the network efficiently. Simulations show that SnFTree provides extra provisioning flexibility and hence outperforms the state-of-the-art scheduling methods in terms of blocking probability and the number of OpenFlow Agent updates.

The time- and space-varying nature of residual bandwidth in inter-datacenter networks (inter-DCNs) makes conventional end-to-end connections difficult to fully utilize the residual bandwidth. A promising solution is to introduce datacenter storage into data-plane paths. Delay-tolerant bulk data can be temporarily stored at intermediate sites and forwarded (SnF) at a later time when inter-DCN is less congested. However, the conventional methods attempt to involve multi-dimensional state information of the entire network in scheduling, which results in high computational complexity. In this paper, our studies reveal that there exist redundant states in scheduling, which cannot provide any performance benefit while imposing extra computational burden. Inspired by this finding, we propose a state-merging scheduling (SMS) method. By merging the state information of the pre-selected links, the SMS method naturally reduces the redundant states involved, which greatly improves the efficiency of SnF scheduling. Simulations demonstrate that the SMS method can outperform the conventional scheduling method, given a limit of the computational cost.

Existing data center networks (DCNs) based on only electronic packet or all-optical switching still pose an exponential increase in power consumption and cost due to the current high demand for digital data and sustainability issues. The recent development of hybrid DCN prototypes is a promising solution offering relatively higher data throughput, low latency, reduction in cost and energy consumption. This paper explores the frontier of hybrid DCN with a special focus on energy consumption and cost. We evaluate the energy consumption per bit and the greenhouse gas (GHG) emission per year of three hybrid switching systems as compared with an optical point to point (ptp) network, results show that the hybrid switching systems will consume less energy per bit and are likely to emit less GHG annually. We present feasibility analysis on the energy consumption and cost of some hybrid DCN prototypes. Evaluation results show that Helios-like and Hybrid Optical Switching-like prototypes achieve a power usage effectiveness (PUE) value lower than 1.2, an index which represents a very efficient level of energy performance in a data center network.

Increasingly bulk data transfers impose an unprecedented challenge on inter-datacenter (DC) wide area networks (WANs). To overcome this challenge, DC storage is exploited and introduced into the data-plane path so that delay-tolerant bulk data can be temporarily stored and forwarded (SnF) at a later time when network is less congested. However, the use of storage introduces additional complexity into the transfer process. In this paper, we evaluate the computational complexity of the SNF scheduling problem. Studies show that desirable performance can be attained by considering only a few alternate routes rather than dynamically routing over the entire network topology. Inspired by this finding, we propose a time-space decoupled (TSD) SNF scheduling method. By decoupling the problem into its spatial and temporal components and solving these components separately, the TSD method obtains good performance while maintaining low complexity. Simulations show that the TSD method can outperform the conventional joint method when the traffic load is medium or higher.

Rapid growth of traffic leads to bottlenecks on inter-domain links, and the residual bandwidth on such links exhibits temporal and spatial variations. There may not be sufficient bandwidth for end-to-end cross domain transmission upon the arrival of bulk data requests. Storage can be introduced into the forwarding path so that the delay tolerant data can be temporarily stored and forwarded when network links become less congested. In this paper, we study the Store-and-Forward (SnF) problem in a multi-domain environment. We present a routing scheme based on a framework named TS-MLG (time-shifted multilayer graph) to perform inter-domain bulk data transfer by taking both spatial and temporal aspects into consideration. With this scheme, the cross-domain routing path, as well as the storage points and the time of transmission, can be determined through conventional routing computations. An updating strategy is proposed to realize temporal bandwidth information exchange between physical and logical layers. Numerical simulations reveal that SnF can improve the inter-domain bandwidth utilization and optimize the network performance at the practical cost of network states maintenance overhead and computational complexity. Since the multi-domain routing is sensitive to the burden of the control plane, we also studied factors that affect computational complexity.

A digital image encryption algorithm based on dynamic deoxyribonucleic acid coding and chaotic operations using hyper digital chaos in frequency‐domain is proposed and demonstrated, where both the amplitude and phase components in frequency‐domain are diffused and scrambled. The proposed encryption algorithm is evaluated through various evaluations of key parameters such as histogram uniformity, entropy, and correlation. Excellent performance of the encrypted image is achieved to resist the statistical attacks, which implies that the statistical properties of the original image are completely destroyed. In the encryption procedure, each cipher pixel is affected by all of the plain‐pixels as well as cipher‐pixels, due to the implementation of chaotic diffusion and scrambling operations, which increases the sensitivity of the encrypted image to the plain‐text, and improves the security against any differential attacks. Moreover, due to the high sensitivity introduced by the hyper digital chaos, a huge key space is provided for the encrypted image to ensure the high security level, thus the encryption algorithm has a strong secure capability against the brute‐force attacks.

Time division multiplexing passive optical networks (TDM-PONs) technologies are viewed as an attractive solution for flexible and cost-effective mobile front-haul for dense deployment of small cells in cloud radio access network (C-RAN) architecture. Due to the high latency of upstream transmission in TDM-PON because of using a dynamic bandwidth allocation (DBA) mechanism to share the upstream bandwidth, it is a challenge for TDM-PON to meet the strict latency requirement of C-RAN mobile front-haul. Several DBA mechanisms have been proposed in the literature to address this issue for IEEE Ethernet passive optical network (i.e. 10G-EPON) based mobile front-haul. However, ITU TDM-PON such as XG-PON have not yet even been explored in the context of mobile front-haul. In this paper, we present an optimized XG-PON-compliant DBA mechanism called Optimized-Round Robin (Optimized-RR) to support front-haul traffic transport over XG-PON in virtualized small-cell C-RAN architecture. We evaluate its performance in terms of delay, jitter and packet loss over a dynamic data rate mobile front-haul traffic by comparing it with two other recently proposed XG-PON- compliant DBAs namely, Group Assured GIANT (g GIANT) and simple Round-Robin (RR-DBA) DBAs. The performance evaluation results not only show its significant improvement in terms of upstream delay and utilization, but also show a lower packet loss and jitter for aggregated small cells front-haul traffic when comparing it to gGIANT and RR-DBA.

Along with the recent surge in scale expansion of data centers, the interconnection scheme is facing a grave challenge. A huge amount of cables between the switches make the system maintenance and heat dissipation extremely difficult. A promising solution to this problem is using the arrayed waveguide grating (AWG), which can provide a set of wavelength links between its inputs and outputs. However, the scalability of the AWG-based interconnection scheme is restricted by the coherent crosstalk and the wavelength granularity of AWGs. In this paper, we propose a generic modular AWG-based interconnection scheme with scalable wavelength granularity for mega data centers. We first devise a matrix-based method to decompose the AWG into a three-stage network of smaller AWGs, while preserving the nonblocking wavelength routing property of the AWGs. We then introduce the concept of wavelength independency based on the partitioning of the optical connections, such that modular AWGs in the network can reuse the same wavelength set with smaller granularity. We show that the proposed modular AWG-based interconnection network can simplify the cabling complexity of data center networks, while preserving the same function and bandwidth as the original data center network.

This paper explores the scheduling problem of input-queued switches, based on a new algebraic method of edge coloring called complex coloring. The proposed scheduling algorithm possesses three important features inherent from complex coloring: parallelizability, optimality and rearrangeability. Parallelizability makes the algorithm running very fast in a distributed manner, optimality ensures that the algorithm always returns a proper connection pattern with the minimum number of required colors, and rearrangeability allows partially re-scheduling the existing connection patterns if the traffic patterns only changes slightly. The amortized time complexity of the proposed parallel scheduling algorithm, in terms of the time to compute a matching in a timeslot, is O(log N), where N is the switch size. As for the scalability of input-queued switches, due to its low complexity, our algorithm can achieve nearly 100 percent throughput and provide acceptable queuing delay when N is large. Furthermore, the complex coloring method naturally provides an adaptive solution to non-uniform input traffic pattern. Thus, the proposed parallel scheduling algorithm is highly robust in the face of traffic fluctuations.

Increasing bulk data transfers have been overwhelming the Internet. To overcome this, optical circuit-switched (OCS) networks are equipped with assistive storage, so that bulk data that are delay tolerant can be temporarily stored at intermediate nodes and forwarded at later times. But, the use of storage greatly complicates the routing problem, since data storage must be incorporated into routing. This motivates us to simplify this issue by applying slotted operations for the network. Intuitively, the slotted network suffers from degraded network performance due to the inefficient utilization incurred by the slot constraint. However, our simulation shows that when the slot size equals to half the mean duration, the blocking probability is reduced from 0.076 to 8.5×10-5, and the number of network reconfigurations is reduced by a factor of 5, compared to the unslotted case. We reveal that in spite of the inefficient utilization, the slotted operations mitigate bandwidth fragmentation. This suggests in the slotted case, more bandwidth gaps on the links are available for accommodating other requests, and they are aligned with each other in time. Requests hence are delivered with less store-and-forward (SnF) operations being performed. Thus, when the number of SnF allowed for routing each request is limited (in order to reduce the computational complexity of routing), requests are more easily served in the slotted than in the unslotted cases. Our research provides clue for designing scalable slotted OCS networks with assistive storage.

5G backhaul requires high bandwidth and hybrid Radio Frequency/ Free Space Optical (RF/FSO) Communication offers Gbps links. The weather affects availability of both mmW RF and FSO links. Below cloud ceiling, the availability of hybrid RF/FSO link is above 85.7% and it can only be used for deadline-constrained large data transfers. At 17-22km above ground, the bit error rate of FSO links reaches below 10-3and it can be used for delay-sensitive packet transfers. In this work, first, we propose a method to determine the required number of wavelengths and storage size for large data transfers and second, we propose a distributed implementation of automatic repeat request protocol to support 10Gbps bandwidth and millisecond delay for delay-sensitive packet transfers. The simulation results show the following: With link availability of 0.85, blocking rate of 0.05 can be achieved for large data transfers, which indicates 0.05 of the cost of transferring data over 5G dmW spectrum is necessary to guarantee deadline. The distributed implementation of automatic repeat request has similar performance with selective repeat. Both theoretical and simulation results show if load exceeds 0.9 of maximum load allowed by packet error probability, delay increases about 20% drastically.

In wavelength-division-multiplexing (WDM) switches, such as arrayed-waveguide-grating (AWG)-based Clos networks, the supporting of multicast traffic must rise to the challenge of route and wavelength assignment (RWA) problem. In this paper, we study the non-blocking multicast RWA problem in two phases with respect to the cascaded combination of an AWG-based broadcast Clos network, called copy network, and a point-to-point AWG-based Clos network. In phase one, input requests generate broadcast trees in the copy network, and then point-to-point connections are established in the AWG-based Clos network in the second phase. The Clos-type AWG-based multicast networks can be constructed from modular AWGs of smaller sizes with the purpose of minimizing the number of wavelengths required and reducing the tuning range of the wavelength selective converters (WSCs). For solving the multicast RWA problem, we extend the rank-based routing algorithm for traditional space-division broadcast Clos networks such that broadcast trees can also be generated in the WDM copy network in a contention-free manner. However, due to wavelength routing properties of AWGs, the subset of requests input to each subnetwork in the middle stage may not satisfy the precondition of the rank-based RWA algorithm. Nevertheless, we prove that this problem can be solved by cyclically shifting the indices of wavelengths in each subnetwork, which provides the key to recursively route the multicast requests in a non-blocking and contention-free manner in the decomposed AWG-based broadcast Clos network. The time complexity of the proposed multicast RWA algorithm is comparable to that of an AWG-based unicast Clos network.

In this paper, we study the power efficiency and delay performance of the burst mode transmission (BTR) strategy for the IEEE 802.3az energy efficient Ethernet (EEE) protocol. In the BTR strategy, the Ethernet interface goes to sleep once its transmission buffer becomes empty and wakes up as soon as the first arrival has waited for time r or the N-th frame arrives at the interface. Based on the number of arrivals during the vacation time, a new approach is proposed to analyze the M/G/1 queue with vacation times that are governed by the arrival process and the r and N parameters of BTR strategy. Our key idea is to establish the connection between the vacation time and the arrival process to account for their dependency. We first derive the distribution of the number of arrivals during a vacation time based on an event tree of the BTR strategy, from which, we obtain the mean vacation time and the power efficiency. Next, from the condition on the number of arrivals at the end of a vacation period, we derive a generalized P-K formula of the mean delay for EEE systems, and prove that the classical P-K formula of the vacation model is only a special case when the vacation time is independent of the arrival process. Our analysis demonstrates that the r policy and N policy of the BTR strategy are compensating each other. The r policy ensures the frame delay is bounded when the traffic load is light, while the N policy ensures the queue length at the end of vacation times is bounded when the traffic load is heavy. These results, in turn, provide the rules to select appropriate r and N. Our analytical results are confirmed by simulations.

Predictable network performance is critical for cloud applications and can be achieved by providing tenants a dedicated virtual data center (VDC) with bandwidth guarantee. Recently, the extended Hose model was applied to the VDC abstraction to characterize the tradeoff between cost and network performance. The acceptability determination problem of a VDC with heterogeneous bandwidth demand was proved to be NP-complete, even in the simple tree topology. In this paper, we investigate the embedding problem for heterogeneous bandwidth VDC in substrate networks of general topology. The embedding problem involves two coupled sub-problems: virtual machine (VM) placement and multipath route assignment. First, we formulate the route assignment problem with linear programming to minimize the maximum link utilization, and provide K-widest path load-balanced routing with controllable splitting paths. Next, we propose a polynomial-time heuristic algorithm, referred to as the perturbation algorithm, for the VM placement. The perturbation algorithm is congestion-aware as it detects the bandwidth bottlenecks in the placement process and then selectively relocates some assigned VMs to eliminate congestion. Simulation results show that our algorithm performs better in comparison with the existing well-known algorithms: first-fit, next-fit, and greedy, and very close to the exponential-time complexity backtracking algorithm in typical data center network architectures. For the tree substrate network, the perturbation algorithm performs better than the allocation-range algorithm. For the homogeneous bandwidth VDC requests, the perturbation algorithm produces a higher success rate than the recently proposed HVC-ACE algorithm. Therefore, it provides a compromised solution between time complexity and network performance.

Despite the high throughput and low complexity achieved by input scheduling based on Birkhoff-von-Neumann (BvN) decomposition, the performance of the BvN switch becomes less predictable when the input traffic is bursty. In this paper, we propose a deflection-compensated BvN (D-BvN) switch architecture to enhance the quasistatic scheduling based on BvN decomposition. D-BvN switches provide capacity guarantee for virtual circuits (VCs) and deflect bursty traffic when overflow occurs. The deflection scheme is devised to offset the excessive buffer requirement of each VC when input traffic is bursty. The design of our conditional deflection mechanism is based on the fact that it is unlikely that the traffic input to VCs is all bursty at the same time; most likely, some starving VCs have spare capacities when some other VCs are in the overflow state. The proposed algorithm makes full use of the spare capacities of those starving VCs to deflect the overflow traffic to other inputs and provide bandwidth for the deflected traffic to re-access the desired VC. Our analysis and simulation results show that this deflection-compensated mechanism can support BvN switches to achieve close to 100% throughput of offered load even with bursty input traffic, and reduces the average end-to-end delay and delay jitter. Also, our result indicates that the packet out-of-sequence probability due to deflection of overflow traffic is negligible, and thus, only a small re-sequencing buffer is needed at each output port. We also compare D-BvN with the well-established online scheduling algorithm iSLIP, and the result demonstrates that D-BvN outperforms iSLIP in terms of the throughput of offered load when the traffic is non-uniform or the traffic load is not very high.

For large data transfers, the Store-and-Transfer WDM Network (STWN) provides storage and lightpaths jointly. Constrained delays of the transfers restrict the number of waiting requests, thus the storage size can be limited. For dimensioning of STWN, we propose a method to determine the number of wavelengths and storage size required to satisfy the demand. The demand is given as a load matrix with a common deadline and blocking, then we express the load matrix as a base load multiplying an integer matrix consisting of mutually prime elements. The dimensioning process has two steps. 1) We model STWN as TDM network and analyse channels of the TDM system as M/G/1 subsystems. We calculate approximate performance of source-destination pairs and minimize the service rate provisioned for the demand with linear search to find the maximum allowed number of time-slots. 2) With the number of time-slots fixed, we optimize routing and wavelength and time-slot assignment to minimize the number of wavelengths required to accept the integer matrix. In step 1, the approximate performance calculated includes turnout, distribution of delay and storage size. Numerical results show how to drastically decrease the required resources.

Traffic characteristics over space and time constitute an important aspect of cellular networks in consideration of resource provision, traffic engineering and system optimization. Despite recent progress in revealing temporal dynamics and spatial inhomogeneity of cellular traffic, limited knowledge about traffic dependence is gained. One of challenges comes from the absence of sustained observations at a network-wide scale. In this paper, we make an analysis on the week-long traffic generated by a large population of users in a city of China, and model traffic dependence along both space and time dimensions. The evaluation results suggest connections between spatio-temporal dependence of cellular traffic and the organization of human lives. Region differences are observed to impact traffic dependence to a great extent. Additionally, interactive knowledge between space and time enhances traffic prediction with a decrease in root-mean-square error of 2.8%~25.2%. We believe that these achievements will benefit multiple research and development areas such as network deploying and simulation researches.

The three-stage Clos networks remain the most popular solution to many practical switching systems to date. The aim of this paper is to show that the modular structure of Clos networks is invariant with respect to the technological changes. Due to the wavelength routing property of arrayed-waveguide gratings (AWGs), non-blocking and contention-free wavelength-division-multiplexing (WDM) switches require that two calls carried by the same wavelength must be connected by separated links; otherwise, they must be carried by different wavelengths. Thus, in addition to the non-blocking condition, the challenge of the design of AWG-based multistage switching networks is to scale down the wavelength granularity and to reduce the conversion range of tunable wavelength converters (TWCs). We devise a logic scheme to partition the WDM switch network into wavelength autonomous cells and show that the wavelength scalability problem can be solved by recursively reusing similar, but smaller, set of wavelengths in different cells. Furthermore, we prove that the rearrangeably non-blocking (RNB) condition and route assignments in these AWG-based three-stage networks are consistent with that of classical Clos networks. Thus, the optimal AWG-based non-blocking Clos networks also can achieve 100% utilization when all input and output wavelength channels are busy.

Network virtualization supports predicted network performance for applications by providing tenants with a virtual data center in multi-tenant data centers. The Hose model was recently extended for the virtual network abstraction and deployed in an Oktopus system, which offers the trade-off between cost and network performance. Embedding algorithms based on Oktopus model were well studied in the single-root tree topology. In this paper, we investigate congestion-aware allocation of virtual data centers in multipath networks. First, we formulate bandwidth constraints with linear programming, and provide a complete solution with exhaustive searching. Next, to reduce time complexity, we propose a perturbation algorithm to VM placement, which detects the bandwidth bottleneck of the current virtual machine placement and then adjusts the assignment to reduce congestion. The perturbation algorithm is compatible with both load-balanced and single-path routing algorithms. We compare the performance of the exhaustive searching algorithm and perturbation algorithm with load-balanced or single-path routing by simulations. The perturbation algorithm with load-balanced routing performs close to the exhaustive searching algorithm while significantly reduces the time complexity. Therefore, it offers a good tradeoff between time complexity and network performance.

In optical satellite network, the possible optical inter-satellite links (OISLs) candidate for each satellite are usually more than laser communication terminals (LCTs) equipped. This poses the link assignment problem (LAP) of how to assign LCTs of each satellite to establish OISLs with the others. In order to make full use of huge bandwidth provided by LCT, the problem of assigning all satellite LCTs to establish OISLs is modeled as a perfect matching problem in the perfect match model, where a perfect matching is searched over a mixed complete bipartite graph. The regular, random greedy and perfect match model based link assignment schemes are considered. Simulation results show that the average node-to-node distance and LCT utilization of perfect match model based link assignment scheme are better than the regular and random greedy link assignment schemes.

In the High Speed Internet (HSI) service of the Fiber-To-The-Home (FTTH) networks, there are increasingly various applications, such as browsing, video streaming, large downloads and online games. They are competing for the fixed bandwidth on a best-effort basis, and finally resulting in the network congestion and poor quality of experience (QoE). Users want to improve the quality of certain applications. However, today's network service controller (e.g. Broadband Remote Access Server, BRAS) lacks mechanisms to meet the users' desire to enhance the QoE of specific applications. Moreover, BRAS still lacks mechanisms to allocate the bandwidth resources properly for users' different applications according to their “sweet points”. “Sweet points” is a specific bandwidth value. The QoE gets worse quickly when the bandwidth is smaller than the “sweet point”, and keeps the same approximately when the bandwidth is larger than the “sweet point”. In this paper, we proposed a novel BRAS architecture using Software-Defined Networking (SDN) technology, which can improve the user's QoE by adjusting the bandwidth of a specific application to its “sweet point” according to their requirements. To demonstrate the feasibility of our proposed novel BRAS, we built a prototype using SDN to help the user to adjust the bandwidth for the specific application and improve the users' QoE. The experimental results show that users could enhance the QoE of specific applications according to users' preference.

The Ethernet passive optical network (EPON) has recently emerged as the mainstream of broadband access networks. The registration process of EPON, which is defined by the IEEE 802.3av standard, is a multi-point control protocol within the media access control layer. As with other contention-based channel access methods, such as ALOHA and CSMA, stability and delay are critical issues concerning the performances of implementing the protocol on systems with finite channel capacity. In this paper, the registration process of an EPON subscriber, called optical network units (ONUs), is modeled as a discrete-time Markov chain, from which we derive the fundamental throughput equation of EPON that characterizes the registration processes. The solutions of this characteristic equation depend on the maximum waiting time. The aim of our stability analysis is to pinpoint the region of the maximum waiting time that can guarantee a stable registration throughput and a bounded registration delay. For a maximum waiting time selected from the stable region, we obtain the expression of registration delay experienced by an ONU attempting to register. All analytic results presented in this paper were verified by simulations.

In this paper, we propose a mathematical model for calculating the steady state performance of congested networks with large number of TCP flows. Marking probability, average queue length, bandwidth and RTT are connected by a group of nonlinear equations. We transform these equations into a constrained optimization problem, which is then solved by SQP algorithm, a robust method for the optimization with inequality constraints. Numerical results with Matlab and simulation results with NS-2 are given to validate the accuracy of the model and algorithm. We study the relationship between the calculation time and congestion network parameters, and provide a good estimation of calculation time at different network scale. It is helpful for network pricing system.

In Ethernet Passive Optical Networks (EPONs) with offline Dynamic Bandwidth Allocation (DBA) framework, the Optical Line Terminal (OLT) will first collect bandwidth requests from all Optical Network Units (ONUs), and then make bandwidth allocation and scheduling decisions for the shared upstream channel. Due to varying Round-Trip Time (RTT) and grant window sizes, the transmission order of ONUs will greatly affect the mean packet delay. In this paper, we address this grant scheduling problem aiming at minimizing the mean packet delay. We prove several theorems which could determine the ONU to transmit first which will minimize the mean packet delay. And then, by iteratively using these theorems, we propose an Integrated Grant Scheduling (IGS) algorithm to determine the optimal transmission order to minimize the mean packet delay. We then conduct simulations to evaluate the performance of our algorithm, and find that the performance of our algorithm is better compared with other algorithms under various conditions.

Bulk data transfers, which require reliable and efficient transfer of terabits or even petabits of data for data-intensive applications, have been extensively studied. In these applications, usually, a specified amount of data needs to be transferred within strict deadline. Previous researches mainly focus on routing deadline-constrained bulk data transfers to improve the utilization of network resources for guarantying deadline requirements, without considering network failures. This paper proposes novel routing algorithm for deadline-constrained bulk data transfers based on Transfer Failure Probability using continuous-time Markov model. The goal of the novel algorithm is to guarantee the deadline requirement by decreasing the failure probability of data transfer request under network failure settings. Simulation evaluates the performance and demonstrates the effectiveness of our algorithm in terms of deadline violation ratio and service blocking ratio under network failures.

Multi-channel delivery is one competitive solution to reduce channel change time in IPTV. We presented the performance analysis on finite duration multi-channel delivery method in our previous work, when channel changes are sequential and adjacent. In this paper, we develop mathematical models to evaluate the performance of this method when channel change happens randomly. This in fact generalizes the previous model. We find that, in typical setup, a delivery duration of 20 seconds is enough to reduce the channel change time by 47.5%, yet the peak bandwidth increase on carrier's uplink is 66.7%. This indicates that the finite duration multi-channel delivery method can improve viewers' experience effectively when channel changes are random.

The integrated computing system optical network has been viewed as a promising platform to support real-time distributed computing applications. For such a system involving so many heterogeneous computing and network resources, faults seem to be inevitable. Therefore, a fault-tolerant scheme is necessary to improve the availability of the integrated system. However, existing joint task scheduling schemes for real-time distributed computing applications generally do not consider the application availability issues when making scheduling decisions. In this paper, we develop an availability-aware joint task scheduling (AAJTS) scheme, by both taking into account availability improvement and timing requirements. The proposed scheme iteratively enhances the application availability under deadline constraint to protect data communication tasks from network link failures. Extensive simulations demonstrate the effectiveness and the feasibility of the proposed scheme.

In this paper, we study the efficiency of inter-domain state advertising in multi-domain networks, in which the intra-domain topology is substituted with a full-mesh topology of abstract links interconnecting the border nodes of the domain. The advertising trigger rate of individual abstract link is taken as the metric of efficiency. We present an analytical model that reveals the contributions of different physical links to the advertising trigger rate of individual abstract links. According to the model, we propose a state based advertising mechanism, called partial link based advertising (PLA) which can discern and advertise the state variation of individual abstract link by monitoring only a portion of physical links. We formulate the problem of how to optimally selecting physical links to monitor, and provide a heuristic algorithm for link selection. Results show that the PLA performs more efficiently than existing mechanisms in terms of advertising trigger rate and blocking probability.

The optical network based distributed computing system has been thought as a promising technology to support large-scale data-intensive distributed applications. For such a system with so many heterogeneous resources and middlewares involved, faults seem to be inevitable. However, for those applications that need to be finished before the given deadline, a fault in the system will lead to the failure of the application. Therefore, fault-tolerant policy is necessary to improve the performance of the system when faults could happen. In this paper, we address to the fault-tolerant problem for the optical network based distributed computing system. We first propose an overlay approach which applies the existing fault-tolerant policies for distributed computing and optical network. Then we present a joint fault-tolerant policy which takes into account the fault tolerance for computing resource and network resource in the same time. We compare the performances of different polices by simulation. The simulation results show that the joint fault-tolerant policy achieves much better performances compared to overlay approaches.

In this paper, we investigated designing multicast-capable optical cross-connects (MC-OXCs) on a basis of the 4- stage multicast network. Firstly, we derive the sufficient wide- sense nonblocking (WSNB) and rearrangeable nonblocking (RNB) conditions for the 4-stage multicast network with only two (the second and output) stages being multicast-capable. Both WSNB and RNB 4-stage multicast networks need 0(N3/2) crosspoints. Then MC-OXCs empoying the WSNB and RNB 4-stage multicast network are proposed and proven to be power efficient in reducing the total power loss caused by light splitting.

Load-balanced switches have received much attention because they are more scalable than other switch architectures. However, a load-balanced switch has the problem of packet mis-sequencing. In this paper, we propose a dynamic mailbox sharing (DMS) scheme to eliminate the mis-sequencing problem of load-balanced switches only at the cost of a very small increase of delay. The key idea is to keep packets of the same flow in order in the load-balanced switch. The DMS scheme is based on two statistical facts in operational networks: the number of simultaneous active flows in the router buffer is far less than that of in-progress flows, and most of the intra-flow packet intervals are longer than the packet delay in the high speed router. In DMS, the packet sequence of the same flow arrived in the input ports is recorded in the mailbox maintained in the output ports. Then, packets of the same flow are delivered according to the order of their arrivals. The mailbox becomes the bottleneck in order to accommodate a large number of flows. We thus propose a dynamic sharing scheme to alleviate the bottleneck and greatly enhance the scalability of the mailbox. By simulations using the real internet traffic traces, we show that with a simple flow splitter mechanism restraining mis-sequencing, the average packet delay using DMS is considerably lower than that of other schemes including uniform frame spreading, padded frame and the CR switch, and it is close to the ideal case without re-sequencing even when the load is very high. The results also demonstrate that the size of mailbox is in the hundreds.

Data-intensive Grid applications require huge data transferring between multiple geographically separated computing nodes where computing tasks are executed. For a future WDM network to efficiently support this type of emerging applications, traditional approaches to establishing lightpaths between given source destination pairs are not sufficient because a computing task may be executed on any one of several computing nodes having the necessary resources. Therefore, lightpath establishment has to be considered jointly with task scheduling to achieve best performance. We study the optimization problems of jointly scheduling both computing resources and network resources. We first present the formulation of two optimization problems with the objectives being the minimization of the completion time of a job and minimization of the resource usage/cost to satisfy a job with a deadline respectively. When the objective is to minimize the completion time, we devise an optimal algorithm for a special type of applications. Furthermore, we propose efficient heuristics to deal with general applications with either optimization objective and demonstrate their good performances via simulation.

It is widely believed that IP over optical networks will be a major component of the next generation Internet However, it is not efficient to map a single multicast IP flow into one light-tree, since the bandwidth of an IP flow required is usually much less than that of a light-tree. In this paper, we study the problem of multicast flow aggregation (MFA) in the IP over optical two-layered networks under the overlay model, which can be defined as follows: given a set of head ends (i.e. optical multicasting sources), each of which can provide a set of contents (i.e. multicast IP flows) with different required transmission bandwidth, and a set of requested content at the access routers (i.e. optical multicasting destinations), find a set of light-trees as well as the optimal aggregation of multicast IP flows in each light-tree. We model MFA by a tri-partite graph with multiple criteria and show that the problem is NP-complete. Optimal solutions are designed by exploiting MFA to formulate an integer linear programming (ILP), with two parameters: the multicast receiving index alpha and the redundant transmitting index beta. We also propose a heuristic algorithm. Finally, we compare the performance of MFA for different combination of alpha and beta via experiments and show our heuristic algorithm is effective for large-scale network in numerical results

To support large-scale IP-TV signal delivering, a new multicast-IP over light-tree network model was proposed in W. Sun et al. In this paper, we study the problem of topology-independent IP group aggregation in the multicast capable optical network. We first use a tri-partite graph to describe the problem, and then formulate the problem by a mixed integer linear programming (MILP) model. Finally, we propose a pre-estimation for the number of light-trees to reduce the searching space, which significantly saves solving time of MILP