In this work we are interested in the problem of scheduling and redistributing data on master-slave platforms. We consider the case were the workers possess initial loads, some of which having to be redistributed in o...
详细信息
This paper reports on the development and evaluation of scheduling algorithms that allow an adequate distribution of applications in opportunistic computational grid elements. This grid can group computing resources o...
详细信息
The efficiency of scheduling algorithms is essential in order to attain optimal performances from parallel programming systems. In this paper we use a portable parallel programming environment we have implemented, the...
详细信息
Advances in networks and the availability of low-cost PCs and workstations has resulted in PC/workstation clusters being used as a computer environment that replaces parallel computers. These clusters recently compris...
详细信息
ISBN:
(纸本)0769507719
Advances in networks and the availability of low-cost PCs and workstations has resulted in PC/workstation clusters being used as a computer environment that replaces parallel computers. These clusters recently comprise different kinds of nodes and are connected over different kinds of networks, such as Ethernet and Myrinet. They are called distributed heterogeneous systems, and efficient collective communication operations for them have been studied. In this paper we propose heuristic scheduling algorithms for efficient gather operations in distributed heterogeneous systems, and cue present the results of simulations evaluating their performance. These results show that gather operations scheduled by these algorithms are more efficient than sequential gather operations in these systems. They also show that, the efficiency of gather operations scheduled by the algorithms depends on the ratio of the cost that depends on message size in point-to-point communication to the cost that does not depend on message size in point-to-point communication.
In this article we consider a set of scheduling problems for wireless data systems, including the single-carrier model motivated by CDMA technologies and the multi-carrier model motivated by OFDM technologies. We surv...
详细信息
ISBN:
(纸本)9781605600864
In this article we consider a set of scheduling problems for wireless data systems, including the single-carrier model motivated by CDMA technologies and the multi-carrier model motivated by OFDM technologies. We survey a number of single-carrier scheduling algorithms together with the assumptions on traffic patterns and channel conditions under which the algorithms perform well. We also present recent work that adapts some of the single-carrier algorithms to the multi-carrier environment. The challenge in adapting these algorithms partially lies in the NP-hard nature of handling resource wastage, which happens when multiple carriers collectively overallocate resources to the same useR. We present solutions based on the greedy approach for maximizing submodular functions over a matroid, We conclude with several open issues in multi-carrier scheduling.
Over the last few years, the usefulness and maturity of artificial intelligence technologies, and in particular of knowledge-based systems, have been demonstrated by the ever growing number of commercial applications ...
详细信息
For cellular wireless networks carrying real-time traffic, it is in the interest of both network operators and customers that throughput guarantees can be offered. In this paper, we formulate an optimization problem w...
详细信息
scheduling algorithms for next generation multi Tbit/sec routers and switches are evaluated. Concurrently, available packet interconnections do not exceed a few hundreds Gbit/sec of aggregate throughput, making traffi...
详细信息
scheduling algorithms for next generation multi Tbit/sec routers and switches are evaluated. Concurrently, available packet interconnections do not exceed a few hundreds Gbit/sec of aggregate throughput, making traffic modeling and computer simulations an extremely important tool. Here, traffic modeling is investigated in terms of arrival process and destination distribution. 'Real world' traffic models are proposed, and extensive simulations are carried out and compared to theoretical predictions. It is shown that conventional scheduling algorithms, such as round robin, perform very well under 'convenient' simulation assumptions, e.g., heavy load and uniform distribution assumptions. However, when traffic modeling is oriented towards modern applications such as bursty arrival process as in compressed video, or non-uniform destination distribution as in Internet sites popularity, detailed and quantitative simulation results prove that algorithm performances are significantly degraded from over 95% to less than 60% of maximum throughput. A modified 'adaptive' round robin algorithm is proposed, yielding significant improvement of 20% in the router's maximum throughput.
We consider the stability region of a mmWave integrated access and backhaul (IAB) network with stochastic arrivals and time-varying link rates. In the scheduling of links, we consider a limit on the number of RF chain...
详细信息
We consider the stability region of a mmWave integrated access and backhaul (IAB) network with stochastic arrivals and time-varying link rates. In the scheduling of links, we consider a limit on the number of RF chains, and the half-duplex constraint which occurs due to the wireless backhaul links. We characterize the stability region, and propose a back-pressure policy for the IAB network under the RF chains and half-duplex constraints. To implement the back-pressure policy, it is required to compute the maximum weighted schedule, which is a complex problem in general. For the IAB network, we present a distributed message passing scheme to compute the maximum weighted schedule, with almost linear complexity. We also investigate a class of local scheduling policies for the IAB network, which have a smaller stability region in general, but require no message passing. We characterize the stability region for the local class, and show that it is same as the global stability region, if the link rates are un-varying. We provide a bound on the gap between local and global regions when the links are time varying. We propose a local max-weight algorithm which achieves the stability region for the local class, and we present numerical results.
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 (Chlamtac and Pinter in IEEE Trans. Comput. C-36(6):729-737, 1987) and 25K in unit disk graphs (http://***/en/ds/***) 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.
暂无评论