See this document in CiteSeerX!

Investigating the Performance of an MPI-IO Implementation on the IBM SP (1998)  (Make Corrections)  (1 citation)
Ay'isha Alexander North Carolina Agriculture Technical State University Dr....



  Home/Search   Context   Related

 
View or download:
ncat.edu/~agents/Publi...parallelio.pdf
Cached:  PS.gz  PS  PDF   Image  Update  Help

From:  ncat.edu/~agents/Publicati...pubs (more)
(Enter author homepages)

Rate this article: (best)
  Comment on this article  
(Enter summary)

Abstract: The results from the performance suite in the ROMIO implementation show the scalability advantage of parallel I/O. To achieve the maximum optimization possible on a large number of processors, programmers must use the advanced features of MPI-IO (i.e., derived datatypes and collective operations). Parallel I/O evaluations were performed to test these different levels. The higher the data access level is, the more complex features it uses. It was found that the total time to perform an... (Update)

Context of citations to this paper:   More

...implemented on the I O system in different ways. There are four different levels (Level 0 Level 3) of expressing access patterns in MPI IO. See [1] for more details. III EXPERIMENTS: ROMIO Benchmarks A set of tests that comes with the ROMIO implementation. These tests...

Cited by:   More
Investigating the Performance of an MPI-IO Implementation .. - Ay'isha Alexander North (1998)   (Correct)

Active bibliography (related documents):   More   All
0.0:   Simplification of Array Access Patterns for Compiler .. - Yunheung Paek, Jay.. (1998)   (Correct)
0.0:   Implementation and Evaluation of Collective I/O in the .. - Rajesh Bordawekar.. (1996)   (Correct)
0.0:   The Globus Project: A Status Report - Foster, Kesselman (1998)   (Correct)

Similar documents based on text:   More   All
0.2:   Data Sieving and Collective I/O in ROMIO - Thakur, Gropp, Lusk (1998)   (Correct)
0.1:   On Implementing MPI-IO Portably and with High Performance - Thakur, Gropp, Lusk (1999)   (Correct)
0.1:   MPI-IO: A Standard, Portable API for High-Performance.. - Thakur, Gropp, Lusk   (Correct)

BibTeX entry:   (Update)

Ay'isha Alexander, Giri Chukkapalli. Investigating the Performance of an MPI-IO Implementation on the IBM SP. August 1998. http://hpc-mccoy.comp.ncat.edu/~ayisha/papers.htm http://citeseer.ist.psu.edu/alexander98investigating.html   More

@misc{ alexander98investigating,
  author = "A. Alexander and G. Chukkapalli",
  title = "Investigating the Performance of an MPI-IO Implementation on the IBM SP",
  text = "Ay'isha Alexander, Giri Chukkapalli. Investigating the Performance of an
    MPI-IO Implementation on the IBM SP. August 1998. http://hpc-mccoy.comp.ncat.edu/~ayisha/papers.htm",
  year = "1998",
  url = "citeseer.ist.psu.edu/alexander98investigating.html" }
Citations (may not include all citations):
142   MPI-2: Extensions to the Message-Passing Interface - Interface - 1997
1   Investigating the Performance of an MPI-IO Implementation on.. - Alexander, Chukkapalli - 1998
1   MPI-IO: NAS Helps Define An Emerging Standard for File Acces.. (context) - Nitezberg - 1996
1   Using MPI-IO for Fast Parallel I/O: A User's Perspective (context) - Van der Wijngaart - 1996
1   MPI-IO: A Parallel File I/O Interface for MPI Version (context) - Wong, Nietzberg et al. - 1995

Documents on the same site (http://www.eng.ncat.edu/~agents/Publications/pubs.html):   More
Fuzzy Versions of Epistemic and Deontic Logic - Gounder, Esterline   (Correct)
Linguistic Understanding with Speech Acts - Thaxton   (Correct)
Motion Planning for Mobile Agents - Hayward Little Senior   (Correct)

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC