Search

Recently, large datasets have become increasingly necessary for most deep learning tasks, however, large datasets may bring some problems, such as disk storage and huge computational expense. Dataset distillation is an emerging field that aims to synthesize a small dataset from the original dataset, then a random model trained on the distillation dataset can achieve comparable performances to the same architecture model trained on the original dataset. Matching Training Trajectories (MTT) achieves a leading performance in this field, but it needs to pre-train 200 expert models before the formal distillation process, which is called buffer process. In this paper, we propose a new method to reduce the consumed time of buffer process. Concretely, we use Mixture of Experts (MoE) to train several expert models parallelly in buffer process. The experiments show our method can achieve a speedup of up to approximately 4 ∼ 8 × in buffer process with getting comparable distillation performances. • Apply Mixture of Experts to accelerate dataset distillation. • Our method is widely effective for trajectories-matching distillation methods. • Experiments show that our method gets comparable performances. [ABSTRACT FROM AUTHOR]

Buffering techniques have been commonly used in mobile video streaming systems to handle the bandwidth fluctuation and to mitigate the impact of the stochastic characteristic of wireless channels on a mobile user’s quality of experience. However, it has been shown by measurement study that users tend to abort when watching videos with mobile devices, which results in a significant wastage of the video data in the buffer. Therefore, one important problem in mobile video streaming is how to manage the buffer at each mobile user. On the other hand, mobile users generally share the wireless media to download the video data, i.e., mobile users compete with each other for the bandwidth to download the video data. Thus, another important problem in mobile video streaming is how to allocate bandwidth among mobile users. In this paper, we propose to optimize the resource management, i.e., to design buffer management strategy at each mobile user and bandwidth allocation strategy among mobile users, for the multiuser mobile video streaming systems. Specifically, we optimize the long-term average total cost of data wastage and quality of experience of mobile users with certain constraints. By introducing virtual queues and employing the Lyapunov optimization theory, we transform the original optimization problem into the drift-plus-penalty minimization problem. Then, we adopt the primal decomposition to decouple the relationship among different mobile users, which decomposes the problem into a master problem with multiple subproblems. A one-dimension full search algorithm is applied to find the global optimal solution to each subproblem, and the subgradient descent algorithm is utilized to update the solution to the master problem. Finally, simulations are conducted to show that the proposed algorithm is effective for the buffer management and bandwidth allocation in a multiuser mobile video streaming system. [ABSTRACT FROM AUTHOR]

Small satellite networks (SSNs) have attracted intensive research interest recently and have been regarded as an emerging architecture to accommodate the ever-increasing space data transmission demand. However, the limited number of on-board transceivers restricts the number of feasible contacts (i.e., an opportunity to transmit data over a communication link), which can be established concurrently by a satellite for data scheduling. Furthermore, limited battery space, storage space, and stochastic data arrivals can further exacerbate the difficulty of the efficient data scheduling design to well match the limited network resources and random data demands, so as to the long-term payoff. Based on the above motivation and specific characteristics of SSNs, in this paper, we extend the traditional dynamic programming algorithms and propose a finite-embedded-infinite two-level dynamic programming framework for optimal data scheduling under a stochastic data arrival SSN environment with joint consideration of contact selection, battery management, and buffer management while taking into account the impact of current decisions on the infinite future. We further formulate this stochastic data scheduling optimization problem as an infinite-horizon discrete Markov decision process (MDP) and propose a joint forward and backward induction algorithm framework to achieve the optimal solution of the infinite MDP. Simulations have been conducted to demonstrate the significant gains of the proposed algorithms in the amount of downloaded data and to evaluate the impact of various network parameters on the algorithm performance. [ABSTRACT FROM AUTHOR]

This paper presents two main contributions toward efficient very large scale integration circuits mapped with simple cells: 1) a complete synthesis flow to provide good-quality circuits mapped only with simple cells; and 2) an area-oriented, level-aware buffering algorithm based on inverter trees to fix cell fanout violations. An effective pattern-based algorithm to identify XORs/XNORs on and-inverter graphs is also presented. We show that efficient implementations in terms of inverter count, transistor count, area, power, and delay can be generated from circuits with a reduced number of both simple cells and inverters, combined with XOR/XNOR-based optimizations. The proposed buffering algorithm can handle all unfeasible fanout occurrences, while: 1) optimizing the number of added inverters and 2) assigning cells to the inverter tree based on their level criticality. When comparing with academic and commercial approaches, we are able to simultaneously reduce the average number of inverters, transistors, area, power dissipation, and delay up to 48%, 4%, 8%, 14%, and 16%, respectively. [ABSTRACT FROM AUTHOR]

In this paper, storage efficient caching based on time domain buffer sharing is considered. The caching policy allows a user’s device to determine whether and how long it should cache a content item according to the prediction of the user’s random request time, also referred to as the request delay information (RDI). In particular, the aim is to maximize the caching gain for communications while limiting its storage cost. To achieve this goal, a queueing theoretic model for caching with infinite buffers is first formulated, in which Little’s law is adopted to obtain the tradeoff between the hit ratio and the average buffer consumption. When there are multiple content classes with different RDIs, the storage efficiency is further optimized by carefully allocating the storage cost. For more practical finite-buffer caching, a $G/GI/L/0$ queue model is formulated, in which a diffusion approximation and the Erlang-B formula are adopted to determine the buffer overflow probability and the corresponding hit ratio. The optimal hit ratio is shown to be limited by the demand probability and buffer size for large and small buffers respectively. In practice, a user may exploit probabilistic caching with random maximum caching time and arithmetic caching without any need for content arrival statistics to efficiently harvest content files from the air. [ABSTRACT FROM AUTHOR]

We study a single server queueing model with admission control and retrials. In the heavy traffic limit, the main queue and retrial queue lengths jointly converge to a degenerate two-dimensional diffusion process. When this model is considered with holding and rejection costs, formal limits lead to a free boundary curve that determines a threshold on the main queue length as a function of the retrial queue length, above which arrivals must be rejected. However, it is known to be a notoriously difficult problem to characterize this curve. We aim instead at optimizing the threshold on the main queue length independently of the retrial queue length. Our main result shows that in the small and large retrial rate limits, this problem is governed by the Harrison-Taksar free boundary problem, which is a Bellman equation in which the free boundary consists of a single point. We derive the asymptotically optimal buffer size in these two extreme cases, as the scaling parameter and the retrial rate approach their limits. [ABSTRACT FROM AUTHOR]

Field programmable gate array (FPGA) has been one of the most popular platforms to implement convolutional neural networks (CNNs) due to its high performance and cost efficiency; however, limited by the on-chip resources, the existing single-FPGA architectures cannot fully exploit the parallelism in CNNs. In this paper, we explore heterogeneous FPGA-based designs to effectively leverage both task and data parallelism, such that the resultant system can achieve the minimum cost while satisfying timing constraints. In order to maximize the task parallelism, we investigate two critical problems: 1) buffer placement, where to place buffers to partition CNNs into pipeline stages and 2) task assignment, what type of FPGA to implement different CNN layers. We first formulate the system-level optimization problem with a mixed integer linear programming model. Then, we propose an efficient dynamic programming algorithm to obtain the optimal solutions. On top of that, we devise an efficient algorithm that exploits data parallelism within CNN layers to further improve cost efficiency. Evaluations on well-known CNNs demonstrate that the proposed techniques can obtain an average of 30.82% reduction in system cost under the same timing constraint, and an average of 1.5 times speedup in performance under the same cost budget, compared with the state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the buffer-aided two-way wireless energy harvesting (EH) relay network comprising two users and one relay, where two users exchange information with the help of the relay via three phases ofEH, multiple-access, and broadcast. By transforming the opportunistic scheduling design problem into an equivalent convex problem, we present the adaptive design for the buffer-aided two-way EH relay network to maximize the long-term achievable rate region. To fulfill the delay sensitive transmission requirements, a delay-aware adaptive transmission (DAAT) scheduling scheme is proposed to guarantee the average end-to-end delivery delay by employing the Lyapunov optimization framework. Our analysis discloses that the average achievable rate region of the two-way wireless EH relay network can be improved when fully considering the potentials by deploying data buffer and energy storage at the relay. There exists an inherent tradeoff among the achievable sum rate, the delivery delay, and the power consumption. It is shown that, when a certain time delay is tolerable, the DAAT scheduling scheme is able to realize a rate region arbitrarily close to the achievable rate region of the buffer-aided two-way wireless EH relay network. [ABSTRACT FROM AUTHOR]

In this paper, we aim to minimize the average file transmission delay via bandwidth allocation and cache placement in two-tier heterogeneous networks with limited storage capacity, which consists of cache capacity and buffer capacity. For average delay minimization problem with fixed bandwidth allocation, although this problem is nonconvex, the optimal solution is obtained in closed form by comparing all locally optimal solutions calculated from solving the Karush–Kuhn–Tucker conditions. To jointly optimize bandwidth allocation and cache placement, the optimal bandwidth allocation is first derived and then substituted into the original problem. The structure of the optimal caching strategy is presented, which shows that it is optimal to cache the files with high popularity instead of the files with big size. Based on this optimal structure, we propose an iterative algorithm with low complexity to obtain a suboptimal solution, where the closed-from expression is obtained in each step. Numerical results show the superiority of our solution compared with the conventional cache strategy without considering cache and buffer tradeoff in terms of delay. [ABSTRACT FROM AUTHOR]

The manufacturers of NAND flash-based solid-state drives (SSDs) are increasing capacity and throughput by enlarging their page size, which is the minimum I/O unit in the NAND flash chips. Because the host and NAND flash chips have different I/O granularity units, the number of subpage requests increases. However, these subpage requests, especially writes, can cause internal fragmentation and endurance problems. Furthermore, subpage write requests inevitably involve read-modify-write (RMW) operations that increase the write response time because of the out-place-update feature in the NAND flash chips. In this paper, we propose a subpage-aware SSD to increase the lifetime and performance by reducing the number of NAND writes and eliminating unnecessary RMW operations. Our scheme attempts to merge subpage write requests to full page write requests in the write buffer to reduce the number of NAND writes and adds size information to the mapping table to detect unnecessary RMW operations. Our proposed scheme reduces the number of NAND writes by up to 30 and 19 percent on average and the write response time by up to 22 and 13 percent on average. [ABSTRACT FROM AUTHOR]

Adaptive streaming allows for dynamic adaptation of the bitrate to varying network conditions, to guarantee the best user experience. Adaptive bitrate algorithms face a significant challenge in correctly estimating the throughput, as the throughput varies widely over time. The current throughput estimation methods cannot distinguish between throughput fluctuations of different amplitude and frequency. In this paper, we propose a throughput estimation method that accurately estimates the throughput based on previous throughput samples. It is robust to short term and small fluctuations, and sensitive to large fluctuations in throughput. Furthermore, we propose a rate adaptive algorithm for video on demand (VoD) services that selects the quality of the video based on the estimated throughput and playback buffer occupancy. The objective of the rate adaptive algorithms is to guarantee high video quality to improve user experience. The proposed algorithm dynamically adjusts the quality level of the video stream. The proposed method selects high quality video segments, while minimizing the risk of playback interruption. Furthermore, the proposed method minimizes the frequency of video rate changes. We show that the algorithm smoothly switches the video rate to improve user experience. Furthermore, we determine that it efficiently utilizes network resources to achieve a high video rate; competing HTTP clients achieve equitable video rates. We also confirm that variations in the playback buffer size and segment duration do not affect the performance of the proposed algorithm. [ABSTRACT FROM AUTHOR]

A method for evaluation of the performance of the disk subsystem of a data buffering and transmission system and results of experimental studies of the performance of the disk subsystem according to the developed method are presented. A conclusion is drawn on the reasonability of the use of hard-disk drives that provide the required read/write rates with a sufficiently high level of reliability in prospective development of the modern high-capacity NearLine SAS (NL-SAS). [ABSTRACT FROM AUTHOR]

In a solar field, the steam production rate and pressure is not constant, but varies and becomes maximum at solar noon. However solar thermal applications such as multi effect desalination with thermo vapour compressor (MED-TVC) system are usually designed for a particular pressure and flow rate. Hence a buffer storage system is needed in solar MED-TVC system to adjust the differences between the steam production and the consumption rates. In the present work steam accumulators were used as buffer storage in direct steam generation (DSG) for correcting the mismatch between production and consumption and also for energy storage. A parametric analysis was performed to investigate the impact of steam accumulator volume and also the irradiance disturbance in solar field on MED-TVC system's performance. This paper presents the transient simulation results of the steam accumulator's output and the corresponding output of the MED-TVC yield from 9:00 h to 20:00 h in a day. [ABSTRACT FROM AUTHOR]

With the rapid progress of mobile Internet, the streaming video service has boom over wireless networks in recent years. A smooth playback experience is crucial for the popularization of these services. However, limited by fluctuating bandwidth and various network impairments, the streaming video inevitably suffers kinds of stalling events, which significantly distorts its temporal structures and results in annoying jerky playback. In this paper, we propose an efficient quality metric to blindly evaluate the user experience for stalled streaming video without using its original sequence. Instead of requiring buffer or manifest information like existing methods, we only access to the decoded video and extract two complementary image features, i.e., global intensity and local texture, to estimate the stall number and duration. Then, by means of a straightforward and easy-to-use linear combination model, we can map the normalized stall number and duration information to a quantitative quality score. Experimental results on the publicly available LIVE-Avvasi mobile video database show that our predicted video quality is highly consistent with the user experience and outperforms many existing quality-of-experience models. [ABSTRACT FROM AUTHOR]

Spatiotemporal proximity analysis to determine spatiotemporal proximal paths is a critical step for many movement analysis methods. However, few effective methods have been developed in the literature for spatiotemporal proximity analysis of movement data. Therefore, this study proposes a space-time-integrated approach for spatiotemporal proximal analysis considering space and time dimensions simultaneously. The proposed approach is based on space-time buffering, which is a natural extension of conventional spatial buffering operation to space and time dimensions. Given a space-time path and spatial tolerance, space-time buffering constructs a space-time region by continuously generating spatial buffers for any location along the space-time path. The constructed space-time region can delimit all space-time locations whose spatial distances to the target trajectory are less than a given tolerance. Five space-time overlapping operations based on this space-time buffering are proposed to retrieve all spatiotemporal proximal trajectories to the target space-time path, in terms of different spatiotemporal proximity metrics of space-time paths, such as Fréchet distance and longest common subsequence. The proposed approach is extended to analyze space-time paths constrained in road networks. The compressed linear reference technique is adopted to implement the proposed approach for spatiotemporal proximity analysis in large movement datasets. A case study using real-world movement data verifies that the proposed approach can efficiently retrieve spatiotemporal proximal paths constrained in road networks from a large movement database, and has significant computational advantage over conventional space-time separated approaches. [ABSTRACT FROM AUTHOR]

Vehicular content networks (VCNs), which distribute medium-volume contents to vehicles in a fully distributed manner, represent the key enabling technology of vehicular infotainment applications. In VCNs, the road-side units (RSUs) cache replicas of contents on the edge of networks to facilitate the timely content delivery to driving-through vehicles when requested. However, due to the limited storage at RSUs and soaring content size for distribution, RSUs can only selectively cache content replicas. The edge caching scheme in RSUs, therefore, becomes a fundamental issue in VCNs. This paper addresses the issue by developing an edge caching scheme in RSUs. Specifically, we first analyze the features of vehicular content requests based on the content access pattern, vehicle's velocity, and road traffic density. A model is then proposed to determine whether and where to obtain the replica of content when the moving vehicle requests it. After this, a cross-entropy-based dynamic content caching scheme is proposed accordingly to cache the contents at the edge of VCNs based on the requests of vehicles and the cooperation among RSUs. Finally, the performance of the proposed scheme is evaluated by extensive simulation experiments. [ABSTRACT FROM AUTHOR]

We study an online buffer management problem under the model introduced by Azar and Gilon (2015) [5] recently. Unit-sized packets arrive and are kept in a First-In-First-Out buffer of size B in an online fashion at a network server. Each packet is associated with an arrival time, a value and a processing cycle time in the buffer. The density of a packet is defined to be the ratio of its value to processing time. It is assumed that every packet can be transmitted only after its processing cycle is completed and only the packet at the head of the buffer can be processed. A packet is allowed to be preempted and then discarded from the buffer. But, the value of a packet is attained only if it is successfully transmitted. Under the model, the objective of online buffer management is to maximize the total value of transmitted packets. This model finds applications to packet scheduling in communication networks. In this study, we consider the model with constant density from a theoretical perspective. We first propose some lower bounds for the problem. Azar and Gilon obtained a 4-competitive algorithm for the online buffer management problem for packets with constant density. Here, we present a ( 2 + 1 B − 1 )-competitive algorithm for the case B > 1 as well as its generalization to the multi-buffer model. Moreover, we prove that the competitive ratio of a deterministic online algorithm is at least four when the buffer size is one. We also conduct experiments to demonstrate the superior performance of the proposed online algorithm against the previous approach. [ABSTRACT FROM AUTHOR]

Session types are behavioural types for guaranteeing that concurrent programs are free from basic communication errors. Recent work has shown that asynchronous session subtyping is undecidable. However, since session types have become popular in mainstream programming languages in which asynchronous communication is the norm rather than the exception, it is crucial to detect significant decidable subtyping relations. Previous work considered extremely restrictive fragments in which limitations were imposed to the size of communication buffer (at most 1) or to the possibility to express multiple choices (disallowing them completely in one of the compared types). In this work, for the first time, we show decidability of a fragment that does not impose any limitation on communication buffers and allows both the compared types to include multiple choices for either input or output, thus yielding a fragment which is more significant from an applicability viewpoint. In general, we study the boundary between decidability and undecidability by considering several fragments of subtyping. Notably, we show that subtyping remains undecidable even if restricted to not using output covariance and input contravariance. [ABSTRACT FROM AUTHOR]

Peer-to-peer (P2P) live streaming systems have gained popularity due to the self-scalability property of the P2P overlay networks. In P2P live streaming, peers retrieve stream content from other peers in the system. Therefore, peer selection strategy is a fundamental element to build an overlay which manages the playback delay and startup delay experienced by the peers. In this paper, we propose a peer selection strategy which manages to build a minimum delay overlay using three different stages of overlay construction. In the first stage, the tracker suggests some peers as prospective partners to a new peer. In the second stage, the peer selects its partners out of these peers such that delay is minimized. The third stage is the topology adaptation phase of peers, where peers reposition themselves in the overlay to maintain minimum delay during peer churn. In the proposed peer selection strategy, peers are selected in all the stages based on parameters such as propagation delay, upload capacity, buffering duration and buffering level. The proposed strategy is compared with two existing strategies in the literature: Fast-Mesh (Ren et al. in IEEE Trans Multimed 11: 1446, 2009) and Hybrid live p2p streaming protocol (Hammami et al., 2014) using simulations. Our results show that playback delay and startup delay are reduced significantly with the help of proposed strategy. We demonstrate that the stability of the system also improves during peer churn. [ABSTRACT FROM AUTHOR]

In this paper, we propose an opportunistic cooperation scheme in a two-user buffer-aided cognitive radio network. In each time slot, the secondary user (SU) will either transmit its own packets, or help the primary user to forward packets, based on a priority factor $\alpha$. According to previous cooperation schemes, some time slots will be wasted if the selected buffer is empty. However, under the proposed scheme, these time slots will be handed over to the nonempty buffer's transmission, which increases the average service rates of buffers at the SU. Considering the bursty nature of sources, we characterize the evolution of buffer states by using the Markov chain. By calculating the probability of time slot handing over, the increased service rates can be theoretically analyzed, so as to the improvements on the stable throughput region and the average end-to-end transmission delay. Moreover, we further discuss the throughput and delay tradeoffs between the two users, which indicates that we can control the traffic load by adjusting the priority factor. Simulations validate our analysis, and reveal that the proposed cooperation scheme with time slot handing over offers a far larger throughput while maintaining a shorter transmission delay compared with previous schemes. [ABSTRACT FROM PUBLISHER]

The impact of a finite relay buffer on the throughput of two-way relay is analyzed from a new queueing theoretic point of view. Distinctively from recent Markov model based analyses, the proposed queueing theoretic analysis is able to infer closed-form asymptotic upper bounds for the throughput, shed valuable insights, and point out limitations in the recent analyses. Validated by simulations, our queueing theoretic analysis reveals that the throughput is increasingly insusceptible to the size of the relay buffer, as the buffer enlarges. Moreover, locking the relay in transmitting xored packets can hardly improve the throughput, especially under balanced channel conditions. This is due to the fact that the relay queues stabilize nonempty, and hence, xored packets are forwarded in most cases. [ABSTRACT FROM PUBLISHER]

We consider Persistence , a new online problem concerning optimizing weighted observations in a stream of data when the observer has limited buffer capacity. A stream of weighted items arrive one at a time at the entrance of a buffer with two holding locations. A processor (or observer) can process (observe) an item at the buffer location it chooses, deriving this way the weight of the observed item as profit. The main constraint is that the processor can only move synchronously with the item stream. Persistence is the online problem of scheduling the processor movements through the buffer so that its total derived value is maximized under this constraint. We study the performance of the straight-forward heuristic Threshold , i.e., forcing the processor to “follow” an item through the whole buffer only if its value is above a threshold. We analyze both the optimal offline and Threshold algorithms in the cases where the input stream is either a random permutation, or its items are iid valued. We show that in both cases the competitive ratio achieved by the Threshold algorithm is at least 2/3 when the only statistical knowledge of the items is the median of all possible values. We generalize our results by showing that Threshold, equipped with some minimal statistical advice about the input, achieves competitive ratios in the whole spectrum between 2/3 and 1, following the variation of a newly defined density-like measure of the input. This result is a significant improvement over the case of arbitrary input streams, where we show that no online algorithm can achieve a competitive ratio better than 1/2. [ABSTRACT FROM AUTHOR]

We propose two cooperative secure transmission schemes to protect a two-hop buffer-aided network assisted by an energy harvesting relay. In the first scheme, we assume that the knowledge of the energy harvesting and fading channels states is known in a non-causal manner (offline). In the second scheme, we assume that this knowledge is known in a causal manner (online). For both schemes, we first design an effective link selection policy by taking into account of the transmission efficiency and information security requirements. We then optimally allocate the harvested power at the relay node. For the offline scheme, we maximize the average secrecy rate under the stability constraints of the data queue and the energy queue according to the proposed link selection policy, and design a two-stage iterative algorithm to select the transmission link and allocate relay’s transmit power. In the online scheme, we first model the average secrecy rate maximum problem as a Markov decision process, and then utilize the causal knowledge to select the best transmission link and optimally allocate relay’s transmit power. In addition, the exact and asymptotic closed-form expressions are derived for the ergodic secrecy rate. Numerical results are presented to validate our analysis and demonstrate that the proposed schemes outperform the other buffered-aided secure transmission schemes assisted by the energy harvesting relay in terms of average secrecy rate. [ABSTRACT FROM PUBLISHER]

Delay Tolerant Network (DTN) multi-copy routing protocols are privileged to create and transmit multiple copies of each message that causes congestion and some messages are dropped. This process is known as reactive drop because messages were dropped re-actively to overcome buffer overflows. The existing reactive buffer management policies apply a single metric to drop source, relay and destine messages. Hereby, selection to drop a message is dubious because each message as source, relay or destine may have consumed dissimilar magnitude of network resources. Similarly, DTN has included time to live (ttl) parameter which defines lifetime of message. Hence, when ttl expires then message is automatically destroyed from relay nodes. However, time-to-live (ttl) is not applicable on messages reached at their destinations. Moreover, nodes keep replicating messages till ttl expires even-though large number of messages has already been dispersed. In this paper, we have proposed Priority Queue Based Reactive Buffer Management Policy (PQB-R) for DTN under City Based Environments. The PQB-R classifies buffered messages into source, relay and destine queues. Moreover, separate drop metric has been applied on individual queue. The experiment results prove that proposed PQB-R has reduced number of messages transmissions, message drop and increases delivery ratio. [ABSTRACT FROM AUTHOR]

We consider the reordering buffer problem on a line consisting of n equidistant points. We show that, for any constant δ , an (offline) algorithm that has a buffer ( 1 − δ ) ⋅ k performs worse by a factor of Ω ( log ⁡ n ) than an offline algorithm with buffer k . In particular, this demonstrates that the O ( log ⁡ n ) -competitive online algorithm MovingPartition by Gamzu and Segev (2009) [9] is essentially optimal against any offline algorithm with a slightly larger buffer. [ABSTRACT FROM AUTHOR]

As flash memory becomes popular, flash memory-based solid-state drive (SSD) has been the major storage device. SSD has numerous merits such as high I/O speed, low energy consumption, strong shock resistance and small form factor. Meanwhile, some shortcomings still exist including erase-before-write and different cost for read, write and erase operation. Aiming at efficient buffer management of SSD, this paper proposes a novel approach based on particle swarm optimization (PSO) algorithm. The PSO algorithm is used to estimate the Predict Hot Fitness value of each logical page in the buffer to correctly identify them either as hot or cold by properly reflecting the spatial and temporal locality. The pages predicted as hot are kept in the buffer to maximize the hit ratio and utilization of the SSD buffer. Simulation with real workload shows that the proposed scheme significantly outperforms the representative SSD buffer management schemes in terms of hit ratio and throughput. [ABSTRACT FROM AUTHOR]

Energy harvesting sensor systems typically incorporate energy buffers (e.g., rechargeable batteries and supercapacitors) to accommodate fluctuations in supply. However, the presence of these elements limits the miniaturization of devices. In recent years, researchers have proposed a new paradigm, transient computing, where systems operate directly from the energy harvesting source and allow computation to span across power cycles, without adding energy buffers. Various transient computing approaches have addressed the challenge of power intermittency by retaining the processor's state using non-volatile memory. However, no generic approach has yet been proposed to retain the state of peripherals external to the processing element. This paper proposes RESTOP, flexible middleware which retains the state of multiple external peripherals that are connected to a computing element (i.e., a microcontroller) through protocols such as SPI or I2C. RESTOP acts as an interface between the main application and the peripheral, which keeps a record, at run-time, of the transmitted data in order to restore peripheral configuration after a power interruption. RESTOP is practically implemented and validated using three digitally interfaced peripherals, successfully restoring their configuration after power interruptions, imposing a maximum time overhead of 15% when configuring a peripheral. However, this represents an overhead of only 0.82% during complete execution of our typical sensing application, which is substantially lower than existing approaches. [ABSTRACT FROM AUTHOR]

Keeping the delay experienced by packets while travelling from a source to a destination below certain thresholds is essential to successfully deliver a number of Internet services nowadays. Most of the packet delay can be usually ascribed to the time spent in the many queues encountered by the packet. In this context, the term bufferbloat has been recently coined to denote the uncontrolled growth of the queuing time due, among others, to the excessive size of the buffers and the attitude of TCP to increase the sending rate until a packet is dropped. In this paper, we focus on the queues employed by the traffic control infrastructure and by the network device drivers. Reducing the queuing time due to the former is the objective of a plethora of scheduling algorithms developed in the past years and referred to as Active Queue Management (AQM) algorithms. Conversely, the impact of the additional queuing in the buffer of the network device driver on performance and on the effectiveness of AQM algorithms has instead received much less attention. In this paper, we report the results of an experimental analysis we conducted to gain a better insight into the impact that network device buffers (and their size) have on performance. We also give an in-depth presentation of Dynamic Queue Limits (DQL), an algorithm recently introduced in the Linux kernel to dynamically adapt the size of the buffers held by network device drivers. The experiments we conducted show that DQL not only enables to reduce the queuing time in the network device buffers, which is essential to ensure the effectiveness of AQM algorithms, but also enables to keep latency stable, which is important to reduce the jitter. In order to faithfully reproduce through simulations the dynamics revealed by the experimental study we conducted, we implemented DQL for the popular ns-3 network simulator. In this paper, we describe the design of such implementation and report the results of a simulation study we conducted to show the ability of the simulator to accurately reproduce the experimental results. [ABSTRACT FROM AUTHOR]

Summary As the retransmission mechanism of the Session Initiation Protocol (SIP) significantly increases the amount of incoming traffic and makes an overloaded SIP server unstable, the models created for a stable SIP system cannot be effectively used to evaluate the performance of the overloaded server. To overcome this difficulty, in this paper, we present a fluid-flow model to analyze the priority-based request scheduling mechanism (PRSM) behavior in the overloaded SIP server equipped with an infinite buffer. The PRSM has been proposed in our earlier studies to prevent the overloading of SIP servers, where the original SIP request messages have strict priority over the retransmissions. The performance evaluation of the proposed model for the PRSM and the reference model of the conventional SIP has been reported under various test scenarios. The experiment results of the proposed model verify the effectiveness of the PRSM fluid-flow model as we already demonstrated in our previous studies using our simulation tool. Without loss of generality, the numerical results closely match with the simulation results, indicating that the proposed models can accurately capture the dynamic behavior of a SIP server with the PRSM. [ABSTRACT FROM AUTHOR]

This paper introduces an efficient, low-power, low-area multi-level 2-D discrete Haar wavelet transform (2-D DHWT) architecture. The proposed architecture consists of four add/subtract elements, register bank, three multiplexers and a buffer memory. All or fractions of the N-stage register are used to perform row-column image transposition for every 2-D DHWT decomposition level. A fixed size block memory of $$\frac{{{\varvec{N}}}}{{{\varvec{2}}}}\times \frac{{{\varvec{N}}}}{{{\varvec{2}}}}$$ -sample is used to store low-low frequency band to perform multi-level 2-D DHWT decomposition, except for the first level where no block memory is used. For synthesis results, the proposed architecture outperforms similar architectures in hardware usage, power consumption and speed. Also, it is found that the power delay product of the proposed architecture is less than 1 $$(\hbox {mW} \times \upmu \hbox {s})$$ compared with up to 1.9 $$(\hbox {mW} \times \upmu \hbox {s})$$ for cascaded architecture. Furthermore, an up to 209 MHz processing speed is achieved which enables a three-level 2-D DHWT decomposition of a $$256\times 256$$ -pixel image to be performed within 0.4 ms. [ABSTRACT FROM AUTHOR]

Buffer overflow vulnerability is a kind of consequence in which programmers’ intentions are not implemented correctly. In this paper, a static analysis method based on machine learning is proposed to assist in auditing buffer overflow vulnerabilities. First, an extended code property graph is constructed from the source code to extract seven kinds of static attributes, which are used to describe buffer properties. After embedding these attributes into a vector space, five frequently used machine learning algorithms are employed to classify the functions into suspicious vulnerable functions and secure ones. The five classifiers reached an average recall of 83.5%, average true negative rate of 85.9%, a best recall of 96.6%, and a best true negative rate of 91.4%. Due to the imbalance of the training samples, the average precision of the classifiers is 68.9% and the average F1 score is 75.2%. When the classifiers were applied to a new program, our method could reduce the false positive to 1/12 compared to Flawfinder. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the cross-layer optimization of caching and fast video delivery for enhanced video streaming quality of experience in two-hop relaying networks, where a base station supplies video data to multiple users with the help of relays. Different from conventional systems, each half-duplex relay node is equipped with a cache and a buffer to facilitate joint scheduling of video fetching and delivery. This introduces channel diversity gains and facilitates fast video delivery. In particular, we investigate two-stage caching and delivery control schemes for the minimization of the overall video delivery time. An offline caching and delivery optimization problem, which assumes full knowledge of user requests and channel state information (CSI), is formulated but turns out to be functional and nonconvex. However, we unveil a hidden quasi-convexity and convexity in the two layers of the decomposed problem and, hence, solve the offline problem optimally and efficiently. Moreover, online video delivery control exploiting statistical CSI is investigated under a stochastic dynamic programming (DP) framework. To mitigate the high computational complexity of DP, we further propose a low-complexity online video delivery algorithm, which achieves close-to-optimal performance in the high buffer capacity regime. Simulation results show that our offline and online schemes can significantly reduce the overall video delivery time due to the degrees of freedom enabled by caching and buffering. Besides, an interesting tradeoff between caching and buffering gains in exploiting the diversity of the wireless channel is revealed. [ABSTRACT FROM AUTHOR]

MapReduce is a popular computing model for parallel data processing on large-scale datasets, which can vary from gigabytes to terabytes and petabytes. Though Hadoop MapReduce normally uses Hadoop Distributed File System (HDFS) local file system, it can be configured to use a remote file system. Then, an interesting question is raised: for a given application, which is the best running platform among the different combinations of scale-up and scale-out Hadoop with remote and local file systems. However, there has been no previous research on how different types of applications (e.g., CPU-intensive, data-intensive) with different characteristics (e.g., input data size) can benefit from the different platforms. Thus, in this paper, we conduct a comprehensive performance measurement of different applications on scale-up and scale-out clusters configured with HDFS and a remote file system (i.e., OFS), respectively. We identify and study how different job characteristics (e.g., input data size, the number of file reads/writes, and the amount of computations) affect the performance of different applications on the different platforms. Based on the measurement results, we also propose a performance prediction model to help users select the best platforms that lead to the minimum latency. Our evaluation using a Facebook workload trace demonstrates the effectiveness of our prediction model. This study is expected to provide a guidance for users to choose the best platform to run different applications with different characteristics in the environment that provides both remote and local storage, such as HPC cluster and cloud environment. [ABSTRACT FROM PUBLISHER]

This paper considers a two-machine continuous flow manufacturing system with a buffer of finite capacity, and where the machines are subjected to failures, minimal repairs, and condition-based preventive maintenance. Each machine can degrade into several discrete states, which are characterized by different performance parameters, ranging from perfect functioning to complete shutdown. When a specified degraded state is exited, a preventive maintenance action is performed to restore the machine to one of the higher performance states. We evaluate the impact of preventive maintenance strategies for each machine, in order to determine their impact on production rate and total cost. A methodology is developed to analyze the complex trade-off between the contributions of preventive maintenance and of the buffer to the system performance. It is shown that preventive maintenance should be scheduled not to optimize the performance for each machine individually; its scheduling is best considered from the perspective of the manufacturing system as a whole. The analysis also reveals the importance of considering machine speeds when optimizing the choice of maintenance policy. [ABSTRACT FROM AUTHOR]

Understanding the real achievable performance of mobile ad hoc networks (MANETs) under practical network constraints is of great importance for their applications in future highly heterogeneous wireless network environments. This paper explores, for the first time, the performance modeling for MANETs under a general limited buffer constraint, where each network node maintains a limited source buffer of size $B_s$ to store its locally generated packets and also a limited shared relay buffer of size $B_r$ to store relay packets for other nodes. Based on the Queuing theory and birth-death chain theory, we first develop a general theoretical framework to fully depict the source/relay buffer occupancy process in such a MANET, which applies to any distributed MAC protocol and any mobility model that leads to the uniform distribution of nodes’ locations in steady state. With the help of this framework, we then derive the exact expressions of several key network performance metrics, including achievable throughput, throughput capacity, and expected end-to-end delay. We further conduct case studies under two network scenarios and provide the corresponding theoretical/simulation results to demonstrate the application as well as the efficiency of our theoretical framework. Finally, we present extensive numerical results to illustrate the impacts of buffer constraint on the performance of a buffer-limited MANET. [ABSTRACT FROM PUBLISHER]