Slashing the Disaggregation Tax in Heterogeneous Data Centers with FractOS

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

Beitragende

  • Lluís Vilanova - , Imperial College London (Autor:in)
  • Lina Maudlej - , Technion-Israel Institute of Technology (Autor:in)
  • Shai Bergman - , Technion-Israel Institute of Technology (Autor:in)
  • Till Miemietz - , Barkhausen Intitut (Autor:in)
  • Matthias Hille - , Professur für Betriebssysteme, Distributed Systems (NFoG) (cfaed) (Autor:in)
  • Nils Asmussen - , Barkhausen Intitut (Autor:in)
  • Michael Roitzsch - , Barkhausen Intitut (Autor:in)
  • Hermann Härtig - , Seniorprofessor für Betriebssysteme (Autor:in)
  • Mark Silberstein - , Technion-Israel Institute of Technology (Autor:in)

Abstract

Disaggregated heterogeneous data centers promise higher efficiency, lower total costs of ownership, and more flexibility for data-center operators. However, current software stacks can levy a high tax on application performance. Applications and OSes are designed for systems where local PCIe-connected devices are centrally managed by CPUs, but this centralization introduces unnecessary messages through the shared data-center network in a disaggregated system. We present FractOS, a distributed OS that is designed to minimize the network overheads of disaggregation in heterogeneous data centers. FractOS elevates devices to be first-class citizens, enabling direct peer-to-peer data transfers and task invocations among them, without centralized application and OS control. FractOS achieves this through: (1) new abstractions to express distributed applications across services and disaggregated devices, (2) new mechanisms that enable devices to securely interact with each other and other data-center services, (3) a distributed and isolated OS layer that implements these abstractions and mechanisms, and can run on host CPUs and SmartNICs. Our prototype shows that FractOS accelerates real-world heterogeneous applications by 47%, while reducing their network traffic by 3×.

Details

OriginalspracheEnglisch
TitelEuroSys 2022 - Proceedings of the 17th European Conference on Computer Systems
Herausgeber (Verlag)Association for Computing Machinery, Inc
Seiten352-367
Seitenumfang16
ISBN (elektronisch)9781450391627
ISBN (Print)9781450391627
PublikationsstatusVeröffentlicht - 28 März 2022
Peer-Review-StatusJa

Konferenz

Titel17th European Conference on Computer Systems, EuroSys 2022
Dauer5 April 2022
StadtRennes
LandFrankreich

Externe IDs

Mendeley c0771692-23bd-3d53-ba98-04df0d8f5307
dblp conf/eurosys/VilanovaMBMHARH22

Schlagworte

Forschungsprofillinien der TU Dresden

Schlagwörter

  • Capabilities, Data center, Distributed systems, Operating systems, Resource disaggregation