Understanding and Exploiting Spatial Properties of System Failures on Extreme-Scale HPC Systems
Authors:
- Saurabh GuptaSGSaurabh Gupta
- Devesh TiwariDTDevesh Tiwari
- Christopher JantziCJChristopher Jantzi
- James RogersJRJames Rogers
- Don MaxwellDMDon Maxwell
- 1 June 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15300889,p. 37-44
Abstract
As we approach exascale, the scientific simulations are expected to experience more interruptions due to increased system failures. Designing better HPC resilience techniques requires understanding the key characteristics of system failures on these systems. While temporal properties of system failures on HPC systems have been well-investigated, there is limited understanding about the spatial characteristics of system failures and its impact on the resilience mechanisms. Therefore, we examine the spatial characteristics and behavior of system failures. We investigate the interaction between spatial and temporal characteristics of failures and its implications for system operations and resilience mechanisms on large-scale HPC systems. We show that system failures have "spatial locality" at different granularity in the system, study impact of different failure-types, and investigate the correlation among different failure-types. Finally, we propose a novel scheme that exploits the spatial locality in failures to improve application and system performance. Our evaluation shows that the proposed scheme significantly improves the system performance in a dynamic and production-level HPC system.Keywords
This publication has 17 references indexed in Scilit:
- Understanding GPU errors on large-scale HPC systems and the implications for system design and operation2015
- Lazy Checkpointing: Exploiting Temporal Locality in Failures to Mitigate Checkpointing Overheads on Extreme-Scale Systems2014
- Improving Log-based Field Failure Data Analysis of multi-node computing systems2011
- Failure-aware resource management for high-availability computing clusters with distributed virtual machinesJournal of Parallel and Distributed Computing, 2010
- A Large-Scale Study of Failures in High-Performance Computing SystemsIEEE Transactions on Dependable and Secure Computing, 2009
- What Supercomputers Say: A Study of Five System Logs2007
- Job Failure Analysis and Its Implications in a Large-Scale Production Grid2006
- BlueGene/L Failure Analysis and Prediction Models2006
- Performance Implications of Failures in Large-Scale Cluster Scheduling2005
- Failure data analysis of a large-scale heterogeneous server environment2004