Filters








86,916 Hits in 7.0 sec

Iteration Based Collective I/O Strategy for Parallel I/O Systems

Zhixiang Wang, Xuanhua Shi, Hai Jin, Song Wu, Yong Chen
2014 2014 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing  
Existing collective I/O optimization strategies mainly focus on the I/O phase efficiency and ignore the shuffle cost that may limit the potential of their performance improvement.  ...  MPI collective I/O is a widely used I/O method that helps data-intensive scientific applications gain better I/O performance.  ...  the shuffle phase because of our optimization for the I/O phase.  ... 
doi:10.1109/ccgrid.2014.61 dblp:conf/ccgrid/WangSJWC14 fatcat:yidrtylnnrafpmdl3i7bycygne

LACIO: A New Collective I/O Strategy for Parallel I/O Systems

Yong Chen, Xian-He Sun, Rajeev Thakur, Philip C. Roth, William D. Gropp
2011 2011 IEEE International Parallel & Distributed Processing Symposium  
Collective I/O is widely considered a critical solution that exploits the correlation among I/O accesses from multiple processes of a parallel application and optimizes the I/O performance.  ...  It holds promise in improving the I/O performance for parallel systems.  ...  Fig. 2 shows the strategy of a two-phase collective I/O read for four processes.  ... 
doi:10.1109/ipdps.2011.79 dblp:conf/ipps/ChenSTRG11 fatcat:5x5bfjnnxvhqvnwr55pw274jfu

Remote I/O

Ian Foster, David Kohr, Rakesh Krishnaiyer, Jace Mogill
1997 Proceedings of the fifth workshop on I/O in parallel and distributed systems - IOPADS '97  
We argue instead for a remote I/O paradigm in which programs use familiar parallel I/O interfaces to access remote filesystems.  ...  It addresses performance issues by providing traditional I/O optimizations such as asynchronous operations and through implementation techniques such as buffering and message forwarding to offload communication  ...  Department of Energy, under Contract W-31-109-Eng-38, in support of the multiagency Scalable I/O project.  ... 
doi:10.1145/266220.266222 dblp:conf/iopads/FosterKKM97 fatcat:4z4kja2yvraqfc6h2wlluv7yby

Compiler-directed I/O optimization

M. Kandemir, A. Choudhary
2002 Proceedings 16th International Parallel and Distributed Processing Symposium  
collective I/O, an optimization that allows each processor to do I/O on behalf of others if doing so improves overall performance.  ...  This paper addresses this I/O problem from a compiler's perspective, and presents an I/O optimization strategy based on access pattern and storage form (file layout) detection.  ...  To do this, our compiler employs an I/O optimization technique, called collective I/O [10, 24] .  ... 
doi:10.1109/ipdps.2002.1015492 dblp:conf/ipps/KandemirC02 fatcat:5jppwhstfbh3hopzoweibfougm

Improving Parallel I/O Performance Using Multithreaded Two-Phase I/O with Processor Affinity Management [chapter]

Yuichi Tsujita, Kazumi Yoshinaga, Atsushi Hori, Mikiko Sato, Mitaro Namiki, Yutaka Ishikawa
2014 Lecture Notes in Computer Science  
Two-Phase I/O (2) • Two-Phase I/O  File accesses  Data exchanges • Collective read accesses 1.  ...  Data communications to collect target data MAMHIP13 @ PPAM 2013, September 9, 2013 10 Two-Phase I/O optimization Performance degradation by large number of Independent small I/O accesses Optimization  ... 
doi:10.1007/978-3-642-55224-3_67 fatcat:cjazxneiuned3ng2f5dw7cjp2m

Scaling parallel I/O performance through I/O delegate and caching system

Arifa Nisar, Wei-keng Liao, Alok Choudhary
2008 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis  
To bridge the gap between parallel computational and I/O performance, we propose a portable MPI-IO layer where certain tasks, such as file caching, consistency control, and collective I/O optimization  ...  are delegated to a small set of compute nodes, collectively termed as I/O Delegate nodes.  ...  An example of this is the two-phase I/O technique [7] . Two-phase I/O is used in ROMIO, a popular MPI-IO implementation developed at Argonne National Laboratory [8] .  ... 
doi:10.1109/sc.2008.5214358 dblp:conf/sc/NisarLC08 fatcat:i6zjcsdk4zai3lup6cav56wlji

Diskdirected I/O for MIMD Multiprocessors [chapter]

2009 High Performance Mass Storage and Parallel I/O  
We propose a new technique, disk-directed I/O, to allow the disk servers to determine the flow of data for maximum performance. Our simulations show that tremendous performance gains are possible.  ...  Many scientific applications that run on today's multiprocessors, such as weather forecasting and seismic analysis, are bottlenecked by their file-I/O needs.  ...  Acknowledgements Thanks to Song Bac Toh and Sriram Radhakrishnan for implementing and validating the disk model; to Chris Ruemmler, John Wilkes, and Hewlett Packard Corporation for allowing us to use their  ... 
doi:10.1109/9780470544839.ch35 fatcat:vncne37s6rg4jkpff2w2op27dm

Scalable I/O and analytics

Alok Choudhary, Wei-keng Liao, Kui Gao, Arifa Nisar, Robert Ross, Rajeev Thakur, Robert Latham
2009 Journal of Physics, Conference Series  
We also enumerate key challenges for the I/O systems and discuss ongoing efforts that address these challenges.  ...  In this work we outline our recent achievements and current research for designing scalable I/O software and enabling data analytics in storage systems.  ...  ROMIO's collective I/O implementation is based on the two-phase I/O strategy proposed in [20] , which includes a request redistribution phase and an I/O phase.  ... 
doi:10.1088/1742-6596/180/1/012048 fatcat:q3fkb774vnewdgr7e3dayvl2bq

A Transparent Collective I/O Implementation

Yongen Yu, Jingjin Wu, Zhiling Lan, Douglas H. Rudd, Nickolay Y. Gnedin, Andrey Kravtsov
2013 2013 IEEE 27th International Symposium on Parallel and Distributed Processing  
Collective I/O is an optimization based on a single collective I/O access.  ...  In this study, we develop a user-level library called transparent collective I/O (TCIO) for application developers to easily incorporate collective I/O optimization into their applications.  ...  " function. collective MPI-IO writing the application level buffer, the I/O operations are divided into two phases.  ... 
doi:10.1109/ipdps.2013.36 dblp:conf/ipps/YuWLRGK13 fatcat:ovhhbsfpunca3plwfscrd63vs4

Collective prefetching for parallel I/O systems

Yong Chen, Philip C. Roth
2010 2010 5th Petascale Data Storage Workshop (PDSW '10)  
The preliminary experimental results show that this new collective prefetching strategy holds promise for improving parallel I/O performance.  ...  Data prefetching can be beneficial for improving parallel I/O system performance, but the amount of benefit depends on how efficiently and swiftly prefetches can be done.  ...  Alok Choudary of Northwestern University for their collective caching code.  ... 
doi:10.1109/pdsw.2010.5668089 fatcat:gthme2jfcrajlnrgvxpetwi4ve

Beyond block I/O

Michael Wei, John D. Davis, Ted Wobber, Mahesh Balakrishnan, Dahlia Malkhi
2013 Proceedings of the 6th International Systems and Storage Conference on - SYSTOR '13  
The basic block I/O interface used for interacting with storage devices hasn't changed much in 30 years.  ...  With the advent of very fast I/O devices based on solid-state memory, it becomes increasingly attractive to make many devices directly and concurrently available to many clients.  ...  And we especially thank Chuck Thacker and his team for building the Beehive system.  ... 
doi:10.1145/2485732.2485739 dblp:conf/systor/WeiDWBM13 fatcat:gsxbrkv3pjfk7pgkvbean5rqcq

Cluster I/O with River

Remzi H. Arpaci-Dusseau, Eric Anderson, Noah Treuhaft, David E. Culler, Joseph M. Hellerstein, David Patterson, Kathy Yelick
1999 Proceedings of the sixth workshop on I/O in parallel and distributed systems - IOPADS '99  
We introduce River, a dafa-flow programming environment and l/O substrate for clusters of computers.  ...  River is based on two simple design features: a high-performanccdirniburedqueue, and a storage redundancy mechanism called gradunfed declusrering.  ...  We'd also like to thank Alan Mainwaring for his work on and support of Active Messages; it is the sine quo non of our work. Many thanks to the anonymous reviewen for all of their helpful commc~ts.  ... 
doi:10.1145/301816.301823 dblp:conf/iopads/Arpaci-DusseauATCHPY99 fatcat:agexl2kxdzgnxjbqchyep4rh6u

Evaluating Application I/O Optimization by I/O Forwarding Layers

Jan Christian Meyer, Jørn Amundsen, Xavier Saez
2012 Zenodo  
Rewriting application I/O for performance and scalability on petaflops machines easily becomes a formidable task in terms of man-hour effort.  ...  This whitepaper describes the work on evaluating the IOFSL I/O Forwarding and Scalability Layer on PRACE applications.  ...  For modeling purposes, any deviation form this is classified as I/O.  ... 
doi:10.5281/zenodo.6241946 fatcat:hkxidgccpzfc7ef6yxhzy6jbci

Disk-directed I/O for MIMD multiprocessors

David Kotz
1997 ACM Transactions on Computer Systems  
We propose a new technique, disk-directed I/O, to allow the disk servers to determine the flow of data for maximum performance. Our simulations show that tremendous performance gains are possible.  ...  Many scientific applications that run on today's multiprocessors, such as weather forecasting and seismic analysis, are bottlenecked by their file-I/O needs.  ...  Acknowledgements Thanks to Song Bac Toh and Sriram Radhakrishnan for implementing and validating the disk model; to Chris Ruemmler, John Wilkes, and Hewlett Packard Corporation for allowing us to use their  ... 
doi:10.1145/244764.244766 fatcat:izl7bfzhzvcuxnylwuzv5sdama

High-Performance Techniques for Parallel I/O [chapter]

Avery Ching, Kenin Coloma, Wei-keng Liao, Alok Choudhary, Jianwei Li
2007 Handbook of Parallel Computing  
Figure 1 .5b illustrates the two phase method for collective I/O [TGL99b] , which uses both POSIX I/O and data sieving. This method is referred to as two phase I/O throughout this chapter.  ...  Two phase I/O adds a layer of I/O indirection.  ... 
doi:10.1201/9781420011296.ch35 fatcat:52nfmmdeunflpehwcbulkwhsly
« Previous Showing results 1 — 15 out of 86,916 results