Slashing the Disaggregation Tax in Heterogeneous Data Centers with FractOS

Research output: Contribution to book/conference proceedings/anthology/reportConference contributionContributedpeer-review

Contributors

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

Original languageEnglish
Title of host publicationEuroSys 2022 - Proceedings of the 17th European Conference on Computer Systems
PublisherAssociation for Computing Machinery, Inc
Pages352-367
Number of pages16
ISBN (electronic)9781450391627
ISBN (print)9781450391627
Publication statusPublished - 28 Mar 2022
Peer-reviewedYes

Conference

Title17th European Conference on Computer Systems, EuroSys 2022
Duration5 April 2022
CityRennes
CountryFrance

External IDs

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

Keywords

Research priority areas of TU Dresden

Keywords

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