Modeling and Analysis of Application Interference on Dragonfly+

TL;DR

This paper evaluates how application communication interference affects performance on Dragonfly+ supercomputer networks using simulation, revealing that intra-job interference can severely degrade performance, while certain communication patterns are more resilient.

Contribution

It provides a quantitative analysis of application interference on Dragonfly+ networks using the CODES toolkit, highlighting the impact of different job placement policies and communication patterns.

Findings

Intra-job interference causes significant performance degradation.

Job isolation reduces inter-job interference for certain communication patterns.

Applications with one-to-all communication are resilient to network interference.

Abstract

Dragonfly class of networks are considered as promising interconnects for next-generation supercomputers. While Dragonfly+ networks offer more path diversity than the original Dragonfly design, they are still prone to performance variability due to their hierarchical architecture and resource sharing design. Event-driven network simulators are indispensable tools for navigating complex system design. In this study, we quantitatively evaluate a variety of application communication interactions on a 3,456-node Dragonfly+ system by using the CODES toolkit. This study looks at the impact of communication interference from a user's perspective. Specifically, for a given application submitted by a user, we examine how this application will behave with the existing workload running in the system under different job placement policies. Our simulation study considers hundreds of experiment configurations including four target applications with representative communication patterns under a variety of network traffic conditions. Our study shows that intra-job interference can cause severe performance degradation for communication-intensive applications. Inter-job interference can generally be reduced for applications with one-to-one or one-to-many communication patterns through job isolation. Application with one-to-all communication pattern is resilient to network interference.