The Impact of Process Placement and Oversubscription on Application Performance: A Case Study for Exascale Computing

Please always quote using this URN: urn:nbn:de:0297-zib-53560
  • With the growing number of hardware components and the increasing software complexity in the upcoming exascale computers, system failures will become the norm rather than an exception for long-running applications. Fault-tolerance can be achieved by the creation of checkpoints during the execution of a parallel program. Checkpoint/Restart (C/R) mechanisms allow for both task migration (even if there were no hardware faults) and restarting of tasks after the occurrence of hardware faults. Affected tasks are then migrated to other nodes which may result in unfortunate process placement and/or oversubscription of compute resources. In this paper we analyze the impact of unfortunate process placement and oversubscription of compute resources on the performance and scalability of two typical HPC application workloads, CP2K and MOM5. Results are given for a Cray XC30/40 with Aries dragonfly topology. Our results indicate that unfortunate process placement has only little negative impact while oversubscription substantially degrades the performance. The latter might be only (partially) beneficial when placing multiple applications with different computational characteristics on the same node.

Download full text files

Export metadata

  • Export Bibtex
Author:Florian Wende, Thomas Steinke, Alexander Reinefeld
Document Type:ZIB-Report
Tag:Fault-tolerance; Oversubscription; Process placement
MSC-Classification:00-XX GENERAL
CCS-Classification:A. General Literature
PACS-Classification:00.00.00 GENERAL
Date of first Publication:2015/03/02
Series (Serial Number):ZIB-Report (15-05)
Published in:Appeared in: Proceedings of the 3rd International Conference on Exascale Applications and Software, EASC 2015, pp. 13 - 18

$Rev: 13581 $