Skip to Main Content
High Performance Parallel I/O
book

High Performance Parallel I/O

by Prabhat, Quincey Koziol
October 2014
Intermediate to advanced content levelIntermediate to advanced
434 pages
13h 28m
English
Chapman and Hall/CRC
Content preview from High Performance Parallel I/O
Los Alamos National Laboratory 75
FIGURE 6.5: PLFS N-to-1 application I/O speed-ups. [Image courtesy of John
Bent (LANL).]
6.4 Conclusion
LANL is innovative in its implementation of an HPC infrastructure, as
they are early adopters of parallel file systems, globally shared parallel file
systems, and burst buffers; but this also may lead to future challenges. They
pushed users to use N-to-1 with a lot of self-describing information in their
I/O, and the users have suffered poor performance, but middleware is now
available that makes N-to-1 perform well.
One important part of LANL’s I/O strategy is to be involved in the na-
tional community. Some examples
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Start your free trial

You might also like

Heterogeneous System Architecture

Heterogeneous System Architecture

Wen-mei W. Hwu
Topics in Parallel and Distributed Computing

Topics in Parallel and Distributed Computing

Sushil K Prasad, Anshul Gupta, Arnold L Rosenberg, Alan Sussman, Charles C Weems
Embedded Computing for High Performance

Embedded Computing for High Performance

João Paiva Cardoso, José Figueiredo Coutinho, Pedro C. Diniz

Publisher Resources

ISBN: 9781466582347