Patterns on a Large - Scale Multiprocessor
نویسندگان
چکیده
Rapid increases in the computational speeds of multiprocessors have not been matched by corresponding performance enhancements in the I/O subsystem. To satisfy the large and growing I/O requirements of some parallel scientific applications, we need parallel file systems that can provide high-bandwidth and high-volume data transfer between the I/O subsystem and thousands of processors. Design of such high-performance parallel file systems depends on a thorough grasp of the expected workload. So far there have been no comprehensive usage studies of multiprocessor file systems. Our CHARISMA project intends to fill this void. The first results from our study involve an iPSC/860 at NASA Ames. This paper presents results from a different platform, the CM-5 at the National Center for SupercomputingApplications. The CHARISMA studies are unique because we collect information about every individual read and write request and about the entire mix of applications running on the machines. The results of our trace analysis lead to recommendations for parallel file system design. First, the file system should support efficient concurrent access to many files, and I/O requests from many jobs under varying load conditions. Second, it must efficiently manage large files kept open for long periods. Third, it should expect to see small requests, predominantly sequential access patterns, application-wide synchronous access, no concurrentfile-sharing between jobs, appreciable byte and block sharing between processes within jobs, and strong interprocess locality. Finally, the trace data suggest that node-level write caches and collective I/O request interfaces may be useful in certain environments. ∗This work was supported in part by, the National Science Foundation under grant number CCR-9113170, the National Center for Supercomputing Applications, NASA Ames Research Center under agreement number NCC 2-849, and Thinking Machines Corporation. †Duke University, email: [ap,carla]@cs.duke.edu ‡Dartmouth College, email: [dfk,nils]@cs.dartmouth.edu §M.I.T, email: [email protected], also affiliated with Thinking Machines Corporation
منابع مشابه
Large-scale atmospheric circulation patterns associated with the occurrence of severe storms Zab basin
متن کامل
Performance Analysis of a New Neural Network for Routing in Mesh Interconnection Networks
Routing is one of the basic parts of a message passing multiprocessor system. The routing procedure has a great impact on the efficiency of a system. Neural algorithms that are currently in use for computer networks require a large number of neurons. If a specific topology of a multiprocessor network is considered, the number of neurons can be reduced. In this paper a new recurrent neural ne...
متن کاملAutomatic Embedded Multicore Generation and Evaluation Methodology: a Case Study of a NOC Based 2400-cores on Very Large Scale Emulator
Future generation embedded multicore will be based on hundreds of processors connected through Network on Chip (NOC) . Design productivity of embedded multicore is a major challenge for the semiconductor industry. In this paper, an automatic very large scale NoC design methodology based on FPGA IP is proposed to accelerate the embedded multicore design productivity using very large scale multi-...
متن کاملPerformance Analysis of a New Neural Network for Routing in Mesh Interconnection Networks
Routing is one of the basic parts of a message passing multiprocessor system. The routing procedure has a great impact on the efficiency of a system. Neural algorithms that are currently in use for computer networks require a large number of neurons. If a specific topology of a multiprocessor network is considered, the number of neurons can be reduced. In this paper a new recurrent neural ne...
متن کاملPreex Computations on Symmetric Multiprocessors (preliminary Draft)
We introduce a new optimal preex computation algorithm on linked lists which builds upon the sparse ruling set approach of Reid-Miller and Blelloch. Besides being somewhat simpler and requiring nearly half the number of memory accesses, we can bound our complexity with high probability instead of merely on average. Moreover, whereas Reid-Miller and Blelloch targeted their algorithm for implemen...
متن کاملOn Multiprocessor Scheduling
Large-scale multiprocessor systems offer the potential of cost-effective high-performance computing, but several key technologies—both hardware and software—are necessary before this potential can be realized. This paper examines the state of the art in one such software component: multiprocessor scheduling. After surveying several important papers and recent work in the field, I argue for a mo...
متن کامل