Name
Abstract | ||
|---|---|---|
Mining repeated patterns (often called motifs) in CPU utilization of computers (also called CPU host load) is of fundamental importance. Many recently emerging applications running on high performance computing systems rely on motif discovery for various purposes, including efficient task scheduling, energy saving, etc. In this paper, we propose an efficient motif discovery framework for CPU host load. The framework is elaborately designed to take into account the important properties in host load data. The framework benefits from its ability of on-line discovery and the adaptivity to work with massive data. The experiments are conducted in this paper and the experimental results show that the proposed method is effective and efficient. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1007/s10766-016-0439-0 | International Journal of Parallel Programming |
Keywords | Field | DocType |
Pattern discovery, Motif, Host load, Time series | Supercomputer,Computer science,Scheduling (computing),CPU time,Parallel computing | Journal |
Volume | Issue | ISSN |
45 | 4 | 1573-7640 |
Citations | PageRank | References |
4 | 0.49 | 24 |
Authors | ||
6 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zhuoer Gu | 1 | 7 | 1.89 |
| Ligang He | 2 | 542 | 56.73 |
| Cheng Chang | 3 | 12 | 4.35 |
| Jianhua Sun | 4 | 192 | 25.27 |
| Hao Chen | 5 | 7 | 1.28 |
| Chenlin Huang | 6 | 48 | 8.83 |