Scheduling real time parallel structure on cluster computing
Tóm tắt
Scheduling a large number of high performance computing applications on a cluster-computing environment is a complex task. This becomes more critical in real time systems. Efficient scheduling strategies are critically important to achieving good performance. A cluster scheduler without enough knowledge of the state of the cluster and the scheduled tasks cannot adequately manage the cluster resources. Accordingly, the available processing power of the participating nodes may experience uncontrolled fragmentation. Thus, some of the submitted applications may be rejected due to tasks missing their deadlines. The literature on scheduling real-time task graphs is much less extensive, especially for providing timing guarantees while maximizing the processing power utilization. In this paper, we present a framework for allocating and scheduling real-time applications represented as parallel task graphs on a cluster. We utilize the available processing power on each processor to accommodate as many tasks as possible while satisfying the required deadline of each task. The algorithm also reduces the communication cost among tasks and the possibility of processing power fragmentation.
Từ khóa
#Processor scheduling #Concurrent computing #High performance computing #Real time systems #Knowledge management #Resource management #Power system management #Timing #Clustering algorithms #CostsTài liệu tham khảo
10.1109/ISCC.2000.860714
schopf, 0, Stochastic scheduling, Proceedings of Super-Computing '99
schopf, 2000, Ten steps for super scheduling, Proc of Cluster Forum 5, 202
10.1109/12.30866
sarkar, 1989, Partitioning and Scheduling Parallel Programs for Multiprocessors
kwok, 1999, Parallel program scheduling techniques for networked computer systems, High Performance Cluster Computing
10.1016/0166-5316(89)90050-3
hamscher, 2000, Evaluation of job-scheduling strategies for grid computing, Proc of 1st IEEE/ACM International Workshop on Grid Computing (Grid 2000) at 7th Int Conference on High Performance Computing (HiPC-2000), 191
10.1109/RTTAS.2001.929882
10.1109/71.308533
10.1109/HCW.2000.843757
smith, 1998, Predicting application run times using historical information, Proc of the IPPS/SPDP '98 Workshop on Job Scheduling Strategies for Parallel Processing, 10.1007/BFb0053984
berman, 1996, application-level scheduling on distributed heterogeneous networks, the 1996 ACM/IEEE Conference on Supercomputing, 39, 10.1145/369028.369109
10.1109/ICDCS.1996.507901
ammar, 2001, Real-Time scheduling of tandem tasks preceding graphs on grid computing, Proc of the 1st IEEE Inter Sympos on Signal Processing and Information Technology
feitelson, 1997, Theory and practice in parallel job scheduling, Proc of 1997 Int Parallel Processing Sym, 1291, 1
faerman, 1999, Adaptive performance prediction for distributed data-intensive applications, Proc of the ACM/IEEE Conf on Supercomputing
10.1109/88.218176
10.1145/267658.267736
gehring, 1999, Scheduling a Metacomputer with Uncooperative Subschedulers, Proc of the IPPS Workshop on Job Scheduling Strategies for Parallel Processing, 1659, 10.1007/3-540-47954-6_10
foster, 1999, The Grid Blueprint for a New Computing Infrastructure