Ad-Hoc File Systems At Extreme Scales

Mehmet Soysal () and Achim Streit ()
Additional contact information
Mehmet Soysal: Karlsruhe Institute of Technology, Steinbuch Centre for Computing
Achim Streit: Karlsruhe Institute of Technology, Steinbuch Centre for Computing

A chapter in High Performance Computing in Science and Engineering '19, 2021, pp 537-549 from Springer

Abstract: Abstract This work presents the results of the project with the acronym ADA-FS (Advanced Data Placement via Ad-hoc File Systems at extreme scale). The project which has been approved for the ForHLR II aims to improve I/O performance for highly parallel applications by using distributed on-demand file systems. These temporary file systems are created on the allocated compute nodes, using the node-local disks for the on-demand file system. Through integration into the scheduling system of the supercomputer, it can be requested like any other resource. The research approach contains the design of the file system itself as well as the questions about the right planning strategy for the necessary I/O transfers. In the granted project for the ForHLR II we are investigating the methods on how to integrate the approach into a HPC system. Also, we are evaluating the impact of the on-demand created file systems to running HPC jobs and the applications.

Keywords: file systems; on-demand file systems; wall time; data staging (search for similar items in EconPapers)
Date: 2021
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:sprchp:978-3-030-66792-4_36

Ordering information: This item can be ordered from
http://www.springer.com/9783030667924

DOI: 10.1007/978-3-030-66792-4_36

Access Statistics for this chapter

More chapters in Springer Books from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().