Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.
About 10,000 papers for “Parallel I/O”
Rajeev Thakur, William Gropp · 2007
This paper was the starting point of MPI-IO in that it was the first attempt to exploit this analogy by applying the (then relatively new) MPI concepts for message passing to the realm of parallel I/O.
Cited by 0
David E. Womble, David S. Greenberg · Parallel Computing · 1997
Cited by 0
Hermanns, M.-A., Stephan, M. · JuSER (Forschungszentrum Jülich) · 2008
Cited by 0
A. Choudhary · Journal of Parallel and Distributed Computing · 1993
Cited by 11
R.E. Vears · Elsevier eBooks · 1990
Cited by 0
Szeremi V. · Science and Technology Facilities Council
Cited by 0Open access
Andy Chih-yuan Kahn, Stott Parker, Milos Ercegovac · 1997
of the Thesis Parallel Simulation of Parallel I/O Programs by Andy Chih-Yuan Kahn Master of Science in Computer Science University of California, Los Angeles, 1997 Professor Rajive Bagrodia, Chair Recent trends in computing have shown vast improvements in processor speed. Although this allows computers to solve ever larger problems, the input/output (I/O) subsystem has always been orders of magnitudes slower than the processor. Relative to processor speed, I/O has not shown as much improvement. This results in I/O performance as a major bottleneck. Just as using more processors in parallel can potentially lead to faster processing, I/O can also be done in parallel. Parallel disk subsystems, parallel file systems, and algorithms for collective I/O have been proposed. However, there are many tradeoffs when using these components for parallel I/O. This thesis presents a parallel simulation library which simulates parallel I/O programs based on MPI and its extension, MPI-IO. This l...
Cited by 0
Yarsun Hsu · 2005
Cited by 0
Steven A. Moyer, Vaidy Sunderam · The International Journal of Supercomputer Applications and High Performance Computing · 1995
Metacomputing environments provide a software plat form for implementing parallel applications utilizing networked computing resources. Such programming environments offer a cost-effective alternative to paral lel machines for many classes of problems. Unfortu nately, most metacomputing environments either pro vide no I/O facilities, or serialize all I/O requests and thus limit application performance and scalability. PIOUS is a parallel file system designed to incorporate true parallel I/O into existing metacomputing environ ments. Because PIOUS is itself a parallel application, the functionality of a given metacomputing environ ment is extended without modification. This paper ex amines the PIOUS software architecture and program ming model and presents preliminary results from a prototype PIOUS implementation. Performance results are encouraging, demonstrating the potential of the PIOUS architecture for achieving scalable file system bandwidth in a network computing environment.
Cited by 15
Michael Beynon · 2007
Commodity hardware and operating systems can be used in place of expensive specialty parallel machines. On such Peer-to-Peer systems, high bandwidth I/O is needed for demanding applications. This work attempts to provide insight into what is needed to get acceptable performance from Peer-to-Peer style Parallel I/O using our Alpha Farm cluster of workstations. 1 Introduction A well known problem is the hierarchy that exists in memory systems. Memory is used here to mean the entire range from registers in a CPU, to on-chip instruction and data caches, to main memory, to disks and tapes. The classic trade-offs being price and performance. If price were not an issue, we could use a large number of registers, or use all main memory, etc. If speed did not matter, we could store everything on tape because of its cost-effectiveness. Since neither of these two scenarios are practical, we must find ways to make the speed problem less severe to impact running times less. For the purposes of thi...
Cited by 0
Ron Oldfield, David Kotz · 1998
Scientific applications are increasingly being implemented on massively parallel supercomputers. Many of these applications have intense I/O demands, as well as massive computational requirements. This paper is essentially an annotated bibliography of papers and other sources of information about scientific applications using parallel I/O. It will be updated periodically.
Cited by 0
Bruce Leasure, David J. Kuck, Sergei Gorlatch et al. · 2011
Cited by 0
Prabhat, Koziol, Quincey · 2014
Gain Critical Insight into the Parallel I/O EcosystemParallel I/O is an integral component of modern high performance computing (HPC), especially in storing and processing very large datasets to facilitate scientific discovery. Revealing the state of the art in this field, High Performance Parallel I/O draws on insights from leading practitioners, researchers, software architects, developers, and scientists who shed light on the parallel I/O ecosystem.The first part of the book explains how large-scale HPC facilities scope, configure, and operate systems, with an emphasis on choices of I/O har
Cited by 0
Erik P. DeBenedictis, Juan Miguel Del Rosario · 1992
Describes input/output (I/O) system software that makes I/O on a massively parallel computer as scalable as computing. The analysis is centered around mapping functions, which form an intellectual framework from which to understand parallel I/O. The I/O system not only provides the raw capability of doing parallel I/O, but is also convenient to use and compatible with Unix. These features are in release 3 of the nCUBE system software. This system will transfer data in parallel between scalable programs and/or I/O devices with any number of units. The first test demonstrated heterogeneous programming, or parallel pipes. Results are provided for a parallel pipe using various mapping combinations. The second test demonstrated I/O between parallel programs and secondary storage. Performance figures are given.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
Cited by 55
Luca DIACONESCU · 1997
Cited by 1Open access
David Kotz · 1996
Scienti c applications are increasingly being implemented on massively parallel supercomputers. Many of these applications have intense I/O demands, as well as massive computational requirements. This paper is essentially an annotated bibliography of papers and other sources of information about scienti c applications using parallel I/O. It will be updated periodically. 1
Cited by 11
2009
Applications executing on massively parallel processors (MPPs) often require a high aggregate bandwidth of lowlatency I/O to secondary storage. In many current MPPs, this requirement has been met by supplying internal parallel I/O subsystems that serve as staging areas for data. Typically, the parallel I/O subsystem is composed of I/O nodes that are linked to the same interconnection network that connects the compute nodes. The I/O nodes each manage their own set of disks. The option of increasing the number of I/O nodes together with the number of compute nodes and with the interconnection network allows for a balanced architecture. We explore the issues motivating the selection of this architecture for secondary storage in MPPs. We survey the designs of some recent and current parallel I/O subsystems, and discuss issues of system configuration, reliability, and file systems.
Cited by 0
N. Galbreath, W. Gropp, D. Levine · 2007
We investigate the needs of some massively parallel applications running on distributed-memory parallel computers at Argonne National Laboratory and identify some common parallel I/O operations. For these operations, routines were developed that hide the details of the actual implementation (such as the number of parallel disks) from the application, while providing good performance. An important feature is the ability for the application programmer to specify that a file be accessed either as a high-performance parallel file or as a conventional Unix file, simply by changing the value of a parameter on the file open call. These routines are examples of a parallel I/O abstraction that can enhance development, portability, and performance of I/O operations in applications. Some of the specific issues in their design and implementation in a distributed-memory toolset are discussed. 1 Introduction In order to run Grand Challenge computational science applications on massively parallel pr...
Cited by 0
Laxmikant V. Kalé, Abhinav Bhatelé, Eric J. Bohm et al. · 2011
Cited by 1
2015
Abstract. Bulk Synchronous Parallel ML or BSML is a functional data-parallel language for programming bulk synchronous parallel (BSP) al-gorithms. The execution time can be estimated and dead-locks and in-determinism are avoided. For large scale applications where parallel pro-cessing is helpful and where the total amount of data often exceeds the total main memory available, parallel disk I/O becomes a necessity. We present here a library of I/O features for BSML and its cost model. 1
Cited by 0