skip to main content
10.1145/3127479.3134597acmconferencesArticle/Chapter ViewAbstractPublication PagesmodConference Proceedingsconference-collections
abstract

Building smart memories and high-speed cloud services for the internet of things with derecho

Published: 24 September 2017 Publication History

Abstract

The coming generation of Internet-of-Things (IoT) applications will process massive amounts of incoming data while supporting data mining and online learning. In cases with demanding real-time requirements, such systems behave as smart memories: a high-bandwidth service that captures sensor input, processes it using machine-learning tools, replicates and stores "interesting" data (discarding uninteresting content), updates knowledge models, and triggers urgently-needed responses.
Derecho is a high-throughput library for building smart memories and similar services. At its core Derecho implements atomic multicast (Vertical Paxos) and state machine replication (the classic durable Paxos). Derecho's replicated<T> template defines a replicated type; the corresponding objects are associated with subgroups, which can be sharded into key-value structures. The persistent<T> and volatile<T> storage templates implement version vectors with optional NVM persistence. These support time-indexed access, offering lock-free snapshot isolation that blends temporal precision and causal consistency.
Derecho automates application management, supporting multigroup structures and providing consistent knowledge of the current membership mapping. A query can access data from many shards or subgroups, and consistency is guaranteed without any form of distributed locking. Whereas many systems run consensus on the critical path, Derecho requires consensus only when updating membership.
By leveraging an RDMA data plane and NVM storage, and adopting a novel receiver-side batching technique, Derecho can saturate a 12.5GB RDMA network, sending millions of events per second in each subgroup or shard. In a single subgroup with 2--16 members, through-put peaks at 16 GB/s for large (100MB or more) objects. While key-value subgroups would typically use 2 or 3-member shards, unsharded subgroups could be large. In tests with a 128-member group, Derecho's multicast and Paxos protocols were just 3--5x slower than for a small group, depending on the traffic pattern. With network contention, slow members, or overlapping groups that generate concurrent traffic, Derecho's protocols remain stable and adapt to the available bandwidth.

Cited By

View all
  • (2023)MRTOM: Mostly Reliable Totally Ordered Multicast, a Network Primitive to Offload Distributed Systems2023 IEEE 43rd International Conference on Distributed Computing Systems (ICDCS)10.1109/ICDCS57875.2023.00022(638-648)Online publication date: Jul-2023
  • (2022)Acuerdo: Fast Atomic Broadcast over RDMAProceedings of the 51st International Conference on Parallel Processing10.1145/3545008.3545041(1-11)Online publication date: 29-Aug-2022

Recommendations

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences
SoCC '17: Proceedings of the 2017 Symposium on Cloud Computing
September 2017
672 pages
ISBN:9781450350280
DOI:10.1145/3127479
Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for third-party components of this work must be honored. For all other uses, contact the Owner/Author.

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 24 September 2017

Check for updates

Qualifiers

  • Abstract

Conference

SoCC '17
Sponsor:
SoCC '17: ACM Symposium on Cloud Computing
September 24 - 27, 2017
California, Santa Clara

Acceptance Rates

Overall Acceptance Rate 169 of 722 submissions, 23%

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)6
  • Downloads (Last 6 weeks)2
Reflects downloads up to 17 Feb 2025

Other Metrics

Citations

Cited By

View all
  • (2023)MRTOM: Mostly Reliable Totally Ordered Multicast, a Network Primitive to Offload Distributed Systems2023 IEEE 43rd International Conference on Distributed Computing Systems (ICDCS)10.1109/ICDCS57875.2023.00022(638-648)Online publication date: Jul-2023
  • (2022)Acuerdo: Fast Atomic Broadcast over RDMAProceedings of the 51st International Conference on Parallel Processing10.1145/3545008.3545041(1-11)Online publication date: 29-Aug-2022

View Options

Login options

View options

PDF

View or Download as a PDF file.

PDF

eReader

View online with eReader.

eReader

Figures

Tables

Media

Share

Share

Share this Publication link

Share on social media