Characterizing the Performance of Emerging Deep Learning, Graph, and High Performance Computing Workloads Under Interference
Authors:
- 27 May 2024
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 468-477
Abstract
Throughput-oriented computing via co-running multiple applications in the same machine has been widely adopted to achieve high hardware utilization and energy saving on modern supercomputers and data centers. However, efficiently co-running applications raises new design challenges, mainly because applications with diverse requirements can stress out shared hardware resources (IO, Network and Cache) at various levels. The disparities in resource usage can result in interference, which in turn can lead to unpredictable co-running behaviors. To better understand application interference, prior work provided detailed execution characterization. However, these characterization approaches either emphasize on traditional benchmarks or fall into a single application domain. To address this issue, we study 25 up-to-date applications and benchmarks from various application domains and form 625 consolidation pairs to thoroughly analyze the execution interference caused by application co-running. Moreover, we leverage mini-benchmarks and real applications to pinpoint the provenance of co-running interference in both hardware and software aspects.Keywords
This publication has 25 references indexed in Scilit:
- Potential of a modern vector supercomputer for practical applications: performance evaluation of SX-ACEThe Journal of Supercomputing, 2017
- Cooper: Task Colocation with Cooperative Games2017
- Proxy-Guided Load Balancing of Graph Processing Workloads on Heterogeneous Clusters2016
- A market approach for handling power emergencies in multi-tenant data center2016
- Venice: Exploring server architectures for effective resource sharing2016
- Paying to save: Reducing cost of colocation data center via rewards2015
- iBench: Quantifying interference for datacenter applications2013
- Defensive loop tiling for shared cache2013
- Interference-Aware Program Scheduling for Multicore Processors2013
- A quantitative study of irregular programs on GPUs2012