Abstract
Driven by ever increasing performance demands of compute-intensive applications, supercomputing systems comprise more and more nodes. This growth is a significant burden for fast group communication primitives and also makes those systems more susceptible to failures of individual nodes. In this paper we present a two-phase fault-tolerant scheme for group communication. Using broadcast as an example, we provide a full-spectrum discussion of our approach - from a formal analysis to LogP-based simulations to a message-passing-based implementation running on a large cluster. Ultimately, we are able to reduce the complex problem of reliable and fault-tolerant collective group communication to a graph theoretical renumbering problem. Both, simulations and measurements, show our solution to achieve a latency reduction of 50% with up to six times fewer messages sent in comparison to existing schemes.
Original language | English |
---|---|
Title of host publication | PPoPP 2019 - Proceedings of the 24th Principles and Practice of Parallel Programming |
Publisher | Association for Computing Machinery |
Pages | 287-299 |
Number of pages | 13 |
ISBN (Electronic) | 9781450362252 |
DOIs | |
State | Published - 16 Feb 2019 |
Event | 24th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP 2019 - Washington, United States Duration: 16 Feb 2019 → 20 Feb 2019 |
Publication series
Name | Proceedings of the ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPOPP |
---|
Conference
Conference | 24th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming, PPoPP 2019 |
---|---|
Country/Territory | United States |
City | Washington |
Period | 16/02/19 → 20/02/19 |
Bibliographical note
Publisher Copyright:© 2019 Copyright held by the owner/author(s).
Keywords
- Gossip
- HPC
- LogP model
- Low-latency broadcast
- MPI