Emulation at Very Large Scale with Distem

Tomasz Buchert 1 Emmanuel Jeanvoine 1 Lucas Nussbaum 1
1 ALGORILLE - Algorithms for the Grid
Inria Nancy - Grand Est, LORIA - NSS - Department of Networks, Systems and Services
Abstract : Prospective exascale systems and large-scale cloud infrastructures are composed of dozens of thousands of nodes. Evaluating applications that target such environments is extremely difficult. In this paper, we present an extension of the Distem emulator to allow experimenting on very large scale emulated platforms thanks to the use of a VXLAN overlay network. We demonstrate that Distem is capable of emulating 40,000 virtual nodes on 168 physical nodes, and use the resulting emulated environment to compare two efficient parallel command runners: TakTuk and ClusterShell.
Document type :
Conference papers
Complete list of metadatas
https://hal.inria.fr/hal-00959616
Contributor : Lucas Nussbaum <>
Submitted on : Monday, June 23, 2014 - 5:11:33 PM
Last modification on : Tuesday, December 18, 2018 - 4:26:02 PM
Long-term archiving on : Tuesday, September 23, 2014 - 12:06:38 PM

Files

distem-scale.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00959616, version 2

Collections

INRIA | GRID5000 | UNIV-LORRAINE | LORIA-NSS | LORIA

Citation

Tomasz Buchert, Emmanuel Jeanvoine, Lucas Nussbaum. Emulation at Very Large Scale with Distem. SCALE Challenge, held in conjunction with the 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID), May 2014, Chicago, United States. ⟨hal-00959616v2⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

680

Files downloads

898

Contact


archive-ouverte@inria.fr