Generalizing fine-grained message aggregation

The divergence of application behavior from optimal network usage leads to performance bottlenecks induced by communication. Communication performances are known to worsen when dealing with large quantities of small messages, due to the overhead of envelopes and going through the communication stack multiple times. Prior work has attempted to mitigate this through the aggregration of small messages, but it has only studied the impact for cases where the size of the message is constant and known ahead of time. This thesis explores the applicability of this optimization to variable-sized messages and machines with a large number of cores, analyzing both the theoretical considerations involved and the performance gains achieved in practice. The work is implemented as an update to the Topological Routing and Aggregation Module(TRAM) of the Charm++ parallel programming system.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Narayanan, Venkatasubrahmanian

Contributors dc:contributor

• Kale, Laxmikant

• Fine-grained parallelism
• message aggregation

Statement dc:rights

• Copyright 2020 Venkatasubrahmanian Narayanan

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/108060

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/108060