Towards Efficient Oversubscription: On the Cost and Benefit of Event-Based Communication in MPI

Publikation: Beitrag in Buch/Konferenzbericht/Sammelband/GutachtenBeitrag in KonferenzbandBeigetragenBegutachtung

Abstract

Contemporary HPC systems use batch scheduling of compute jobs running on exclusively assigned hardware resources. During communication, polling for progress is the state of the art as it promises minimal latency. Previous work on oversubscription and event-based communication, i.e. vacating the CPU while waiting for communication to finish, shows that these techniques can improve the overall system utilisation and reduce the energy consumption. Despite these findings, neither of the two techniques is commonly used in HPC systems today. We believe that the current lack of detailed studies of the low-level effects of event-based communication, a key enabler of efficient oversubscription for classical MPI-based applications, is a major obstacle to a wider adoption. We demonstrate that the sched_yield system call, which is often used for oversubscription scenarios, is not best suited for this purpose on modern Linux systems. Furthermore, we incorporate event-based communication into Open MPI and evaluate the effects on latency and energy consumption using an MPI micro-benchmark. Our results indicate that event-base communication incurs significant latency overhead but also saves energy. Both effects grow with the imbalance of the application using MPI.

Details

OriginalspracheEnglisch
Titel2022 IEEE/ACM International Workshop on Runtime and Operating Systems for Supercomputers (ROSS)
Herausgeber (Verlag)IEEE
Seiten1-10
Seitenumfang10
ISBN (elektronisch)978-1-6654-7566-2
ISBN (Print)978-1-6654-7567-9
PublikationsstatusVeröffentlicht - 27 Jan. 2023
Peer-Review-StatusJa

Workshop

TitelInternational Workshop on Runtime and Operating Systems for Supercomputers
KurztitelROSS
Veranstaltungsnummer12
Dauer13 November 2022
Webseite
BekanntheitsgradInternationale Veranstaltung
Ort
StadtDallas
LandUSA/Vereinigte Staaten

Externe IDs

unpaywall 10.1109/ross56639.2022.00001
Scopus 85147799576
dblp conf/ross-ws/BierbaumPH22
Mendeley b4e266b6-852a-377c-9152-2cbb1554ab8a

Schlagworte

Schlagwörter

  • oversubscription, event-based communication, MPI, latency, energy