Loading...
Please wait, while we are loading the content...
Similar Documents
Characterizing the spatio-temporal burstiness of storage workloads
| Content Provider | ACM Digital Library |
|---|---|
| Author | Xu, Gao Mei, Li Mingshu, Li Chen, Zhao Yanjun, Wu |
| Abstract | Computing technology are undergoing evolutionary changes in platform and environment. Computing becomes more and more data-intensive. The design of the data center storage of cloud-based system determines whether data could be accessed efficiently. Characterizing storage workloads can provide valuable information for storage system design on various aspects, such as system modeling, design decision-making, and simulation-based performance evaluation. I/O behaviors in data center environments typically exhibit temporal burstiness and spatial locality. The strong spatio-temporal correlation of I/O behaviors motivates us to study the spatio-temporal burstiness of storage workloads. In this paper, we analyze several I/O traces collected in enterprise storage systems and characterize the spatio-temporal burstiness by abstracting the spatio-temporal correlation of temporal and spatial behaviors. Then a stochastic model is proposed to emulate and predict I/O arrival rates. We conduct an experimental study based on real-world I/O workloads traces. The results demonstrate that abstracting the spatio-temporal correlation of I/O behaviors contributes to improving the accuracy of the storage workloads prediction. |
| Starting Page | 1 |
| Ending Page | 6 |
| Page Count | 6 |
| File Format | |
| ISBN | 9781450334785 |
| DOI | 10.1145/2744210.2744211 |
| Language | English |
| Publisher | Association for Computing Machinery (ACM) |
| Publisher Date | 2015-04-21 |
| Publisher Place | New York |
| Access Restriction | Subscribed |
| Subject Keyword | Burstiness Workload characterization Spatio-temporal correlation Data center storage |
| Content Type | Text |
| Resource Type | Article |
