Skip to main content
USENIX
  • Conferences
  • Students
Sign in
  • Home
  • Attend
    • Registration Information
    • Registration Discounts
    • Venue, Hotel, and Travel
    • Students and Grants
    • Co-located Events
      • SOUPS 2016
      • HotCloud '16
      • HotStorage '16
  • Program
    • At a Glance
    • Technical Sessions
  • Activities
    • Birds-of-a-Feather Sessions
    • Poster Session
  • Participate
    • Instructions for Authors and Speakers
    • Call for Papers
    • Call for Practitioner Talks
  • Sponsorship
  • About
    • Organizers
    • Help Promote!
    • Questions
    • Past Conferences
  • Home
  • Attend
  • Program
  • Activities
  • Participate
  • Sponsorship
  • About

sponsors

Gold Sponsor
Gold Sponsor
Gold Sponsor
Gold Sponsor
Silver Sponsor
Silver Sponsor
Silver Sponsor
Silver Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Media Sponsor
Industry Partner
Industry Partner
Industry Partner

help promote

USENIX ATC '16

Get
Help Promote graphics!

connect with us


  •  Twitter
  •  Facebook
  •  LinkedIn
  •  Google+
  •  YouTube

twitter

Tweets by @usenix

usenix conference policies

  • Event Code of Conduct
  • Conference Network Policy
  • Statement on Environmental Responsibility Policy

You are here

Home ยป Design Guidelines for High Performance RDMA Systems
Tweet

connect with us

Design Guidelines for High Performance RDMA Systems

Authors: 

Anuj Kalia, Carnegie Mellon University; Michael Kaminsky, Intel Labs; David G. Andersen, Carnegie Mellon University

Awarded Best Student Paper

Abstract: 

Modern RDMA hardware offers the potential for exceptional performance, but design choices including which RDMA operations to use and how to use them significantly affect observed performance. This paper lays out guidelines that can be used by system designers to navigate the RDMA design space. Our guidelines emphasize paying attention to low-level details such as individual PCIe transactions and NIC architecture. We empirically demonstrate how these guidelines can be used to improve the performance of RDMA-based systems: we design a networked sequencer that outperforms an existing design by 50x, and improve the CPU effciency of a prior highperformance key-value store by 83%. We also present and evaluate several new RDMA optimizations and pitfalls, and discuss how they affect the design of RDMA systems.

Anuj Kalia, Carnegie Mellon University

Michael Kaminsky, Intel Labs

David G. Andersen, Carnegie Mellon University

Open Access Media

USENIX is committed to Open Access to the research presented at our events. Papers and proceedings are freely available to everyone once the event begins. Any video, audio, and/or slides that are posted after the event are also free and open to everyone. Support USENIX and our commitment to Open Access.

BibTeX
@inproceedings {196243,
author = {Anuj Kalia and Michael Kaminsky and David G. Andersen},
title = {Design Guidelines for High Performance {RDMA} Systems},
booktitle = {2016 USENIX Annual Technical Conference (USENIX ATC 16)},
year = {2016},
isbn = {978-1-931971-30-0},
address = {Denver, CO},
pages = {437--450},
url = {https://www.usenix.org/conference/atc16/technical-sessions/presentation/kalia},
publisher = {USENIX Association},
month = jun,
}
Download
Kalia PDF
View the slides

Presentation Audio

MP3 Download

Download Audio

Award: 
Best Student Paper
  • Log in or    Register to post comments

Gold Sponsors

Silver Sponsors

Media Sponsors & Industry Partners

© USENIX

  • Privacy Policy
  • Contact Us