This paper investigates co-scheduling algorithms for processing a set of parallel applications. Instead of executing each application one by one, using a maximum degree of parallelism for each of them, we aim at sched...
详细信息
This paper investigates co-scheduling algorithms for processing a set of parallel applications. Instead of executing each application one by one, using a maximum degree of parallelism for each of them, we aim at scheduling several applications concurrently. We partition the original application set into a series of packs, which are executed one by one. A pack comprises several applications, each of them with an assigned number of processors, with the constraint that the total number of processors assigned within a pack does not exceed the maximum number of available processors. The objective is to determine a partition into packs, and an assignment of processors to applications, that minimize the sum of the execution times of the packs. We thoroughly study the complexity of this optimization problem, and propose several heuristics that exhibit very good performance on a variety of workloads, whose application execution times model profiles of parallel scientific codes. We show that co-scheduling leads to faster workload completion time (40 % improvement on average over traditional scheduling) and to faster response times (50 % improvement). Hence, co-scheduling increases system throughput and saves energy, leading to significant benefits from both the user and system perspectives.
scheduling MAC-layer transmissions in multi-hop wireless networks is an active and stimulating area of research. There are several interesting algorithms proposed in the literature in the problem space of scheduling f...
详细信息
scheduling MAC-layer transmissions in multi-hop wireless networks is an active and stimulating area of research. There are several interesting algorithms proposed in the literature in the problem space of scheduling for multi-hop wireless networks, specifically for (a) WiMAX mesh networks, (b) long distance multi-hop WiFi networks, and (c) Vehicular Ad-hoc Networks (VANETs). In general, these algorithms have several dimensions in terms of the assumptions made, the input space considered and the solution space generated. In this context, the goal of this survey is three-fold. Firstly, we classify the scheduling algorithms proposed in the literature based on following parameters: problem setting, problem goal, type of inputs and solution technique. Secondly, we describe different scheduling algorithms based on this classification framework. We specifically cover the state-of-the-art scheduling mechanisms proposed for generic multichannel, multi-radio wireless mesh networks and in particular scheduling algorithms for WiMAX mesh networks, long distance mesh networks and vehicular ad-hoc networks. We describe scheduling algorithms which consider scheduling data, voice as well as video traffic. Finally, we compare these algorithms based on our classification parameters. We also critique individual mechanisms and point out the practicality and the limitations, wherever applicable. We observe that, the literature in the domain of scheduling for wireless mesh network is quite extensive, in terms of depth as well as breadth. Our classification framework helps in understanding the pros and cons of various aspects of scheduling for wireless multi-hop (popularly known as wireless mesh) networks. We also list desirable properties of any scheduling mechanism and use our classification framework to point out the open research issues in the space of scheduling for wireless mesh networks.
It is shown that the performance of the maximal scheduling algorithm in wireless ad hoc networks under the hypergraph interference model can be further away from optimal than previously known. The exact worst-case per...
详细信息
It is shown that the performance of the maximal scheduling algorithm in wireless ad hoc networks under the hypergraph interference model can be further away from optimal than previously known. The exact worst-case performance of this distributed, greedy scheduling algorithm is analyzed.
Optimizing downlink coordinated multipoint (CoMP) performance through advanced scheduling algorithms enhances V2X communication technology, enabling efficient resource allocation, minimizing interference, and maximizi...
详细信息
Optimizing downlink coordinated multipoint (CoMP) performance through advanced scheduling algorithms enhances V2X communication technology, enabling efficient resource allocation, minimizing interference, and maximizing data rates for reliable and synchronized communication between vehicles and infrastructure. In this paper, several scheduling algorithms were compared, including support vector machine (SVM) linear, SVM Radial Basis Function (RBF), SVM Sigmoid, Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), and Graph Convolutional Networks (GCN). The performance metrics used in this comparison included CoMP decision, throughput, and cell edge throughput. The results showed that the out-rated GCN algorithm had the best-triggering composition for 5G radio networks, outperforming the other algorithms in terms of CoMP decision accuracy and overall throughput. In particular, the GCN algorithm demonstrated significant improvements in cell edge throughput, which is critical for ensuring reliable communication in areas with weaker signal strength. The reported results proves that the integration of advanced scheduling algorithms in the downlink CoMP framework enhances the efficiency of V2X communication, enabling optimized resource allocation, interference mitigation, and maximized throughput, thereby improving system efficiency, reducing latency, and ensuring reliable and seamless information exchange for connected vehicles, smart cities, and industrial automation.
In this paper, we are interested in using large-deviations theory to characterize the asymptotic decay-rate of the queue-overflow probability for distributed wireless scheduling algorithms, as the overflow threshold a...
详细信息
In this paper, we are interested in using large-deviations theory to characterize the asymptotic decay-rate of the queue-overflow probability for distributed wireless scheduling algorithms, as the overflow threshold approaches infinity. We consider ad hoc wireless networks where each link interferes with a given set of other links, and we focus on a distributed scheduling algorithm called Q-SCHED, which is introduced by Gupta et al. First, we derive a lower bound on the asymptotic decay rate of the queue-overflow probability for Q-SCHED. We then present an upper bound on the decay rate for all possible algorithms operating on the same network. Finally, using these bounds, we are able to conclude that, subject to a given constraint on the asymptotic decay rate of the queue-overflow probability, Q-SCHED can support a provable fraction of the offered loads achievable by any algorithms. (C) 2010 Elsevier B.V. All rights reserved.
The performance of scheduling algorithms for a reservation system is investigated. In this system, a user request is characterised by its start time, resource requirement and holding time. Of interest are scheduling a...
详细信息
The performance of scheduling algorithms for a reservation system is investigated. In this system, a user request is characterised by its start time, resource requirement and holding time. Of interest are scheduling algorithms to handle user requests in a loss system where resource requirements may vary. A Markov decision process formulation is used to obtain the optimal scheduling decisions. Two special cases are considered in depth;they correspond to optimal algorithms that minimise the blocking probability and maximise the channel utilisation, respectively. Analytic results are also obtained for the blocking probability and channel utilisation for an arbitrary scheduling algorithm. Using these results, the performance of first come, first served (FCFS) and the two optimal algorithms is compared. We also prove that FCFS is optimal for maximising channel utilisation when the resource requirement follows a uniform distribution.
scheduling algorithms are important components in the provision of guaranteed quality of service parameters such as delay, delay jitter, packet loss rate, or throughout. The design of scheduling algorithms for mobile ...
详细信息
scheduling algorithms are important components in the provision of guaranteed quality of service parameters such as delay, delay jitter, packet loss rate, or throughout. The design of scheduling algorithms for mobile communication networks is especially challenging given the highly variable link error rates and capacities, and the changing mobile station connectivity typically encountered in such networks. This article provides a survey of scheduling techniques for several types of wireless networks. Some of the challenges in designing such schedulers are first discussed. Desirable features and classifications of schedulers are then reviewed. This is followed by a discussion of several scheduling algorithms which have been proposed for TDMA, CDMA, and multihop packet networks.
In this paper, we have considered the distributed scheduling problem for channel access in TDMA wireless mesh networks. The problem is to assign time-slot(s) for nodes to access the channels, and it is guaranteed that...
详细信息
In this paper, we have considered the distributed scheduling problem for channel access in TDMA wireless mesh networks. The problem is to assign time-slot(s) for nodes to access the channels, and it is guaranteed that nodes can communicate with all their one-hop neighbors in the assigned time-slot(s). And the objective is to minimize the cycle length, i.e., the total number of different time-slots in one scheduling cycle. In single-channel ad hoc networks, the best known result for this problem is proved to be K (2) in arbitrary graphs (IEEE Trans Comput C-36(6):729-737, 1987) and 25K in unit disk graphs (IEEE/ACM Trans Netw pp 166-177, 1993) with K as the maximum node degree. There are multiple channels in wireless mesh networks, and different nodes can use different control channels to reduce congestion on the control channels. In this paper, we have considered two scheduling models for wireless mesh networks. The first model is that each node has two radios, and the scheduling is simultaneously done on the two radios. We have proved that the upper bound of the cycle length in arbitrary graphs can be 2K. The second model is that the time-slots are scheduled for the nodes regardless of the number of radios on them. In this case, we have proved that the upper bound can be (4K-2). We also have proposed greedy algorithms with different criterion. The basic idea of these algorithms is to organize the conflicting nodes by special criterion, such as node identification, node degree, the number of conflicting neighbors, etc. And a node cannot be assigned to a time-slot(s) until all neighbor nodes, which have higher criterion and might conflict with the current node, are assigned time-slot(s) already. All these algorithms are fully distributed and easy to realize. Simulations are also done to verify the performance of these algorithms.
Peer-to-Peer file sharing applications in the Internet, such as BitTorrent, Gnutella, etc., have been immensely popular. Prior research mainly focuses on peer and content discovery, overlay topology formation, fairnes...
详细信息
Peer-to-Peer file sharing applications in the Internet, such as BitTorrent, Gnutella, etc., have been immensely popular. Prior research mainly focuses on peer and content discovery, overlay topology formation, fairness and incentive issues, etc. However, little attention has been paid to investigate the data distribution problem which is also a core component of any file sharing application. In this paper, we present the first effort in addressing this collaborative file distribution problem and formally define the scheduling problem in a simplified context. We develop several algorithms to solve the problem and study their performance. We deduce a theoretical bound on the minimum download time experienced by users and also perform simulations to evaluate our algorithms. Simulation results show that our graph-based dynamically weighted maximum-flow algorithm outperforms all other algorithms. Therefore, we believe our algorithm is a promising solution to be employed as the core scheduling module in P2P file sharing applications.
Analytical expressions for scheduling gain and spectral efficiency of the proportional fair and maximum rate scheduling algorithms for Orthogonal Frequency Division Multiple Access (OFDMA) based systems are derived fo...
详细信息
Analytical expressions for scheduling gain and spectral efficiency of the proportional fair and maximum rate scheduling algorithms for Orthogonal Frequency Division Multiple Access (OFDMA) based systems are derived for the following cases: a) multipath Rayleigh and multipath Nakagami fading, b) Composite channel models which model the combined effect of both small scale and large scale fading. It is shown using Extreme Value Theory (EVT) that the asymptotic distribution of the maxima of the received signal to noise power (SNR) across all the users converges to a Gumbel distribution for both cases. Therefore, we use the Gumbel distribution to derive expressions for both the spectral efficiency and scheduling gain. The scheduling gains obtained through numerical integration (whenever tractable) and simulations match with the analytical values obtained using the EVT based expressions. We also discuss how the moments and order statistics of the Gumbel distribution can be used to study other metrics of the scheduling algorithms.
暂无评论