Abstract
A new dynamic resource allocation algorithm is researched for data center oriented to cloud computing. Applications are modeled using Markov process method,and the change pattern of resource requirements inside the applications is calculated. A resource evaluation approach is proposed based on the expectation of execution waiting time to guarantee application execution performance. An application-aware dynamic resource allocation algorithm is proposed and compared with static max resource allocation algorithm. The experimental results show that,this algorithm reduces the resource consumption by 49% under a same application load,improves the acceptance rate of application request by about 45% for a same given physical server number.
| Translated title of the contribution | Application-aware dynamic resource allocation in data center |
|---|---|
| Original language | Chinese (Traditional) |
| Pages (from-to) | 322-328 |
| Number of pages | 7 |
| Journal | Nanjing Li Gong Daxue Xuebao/Journal of Nanjing University of Science and Technology |
| Volume | 42 |
| Issue number | 3 |
| DOIs | |
| State | Published - 28 Jun 2018 |
| Externally published | Yes |