2017 IEEE 10th International Conference on Cloud Computing (CLOUD) (2017)
Honolulu, Hawaii, United States
June 25, 2017 to June 30, 2017
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CLOUD.2017.12
Nowadays many companies and organizations choose to deploy their applications in data centers to leverage resource sharing. The increase in tasks of multiple applications, however, makes it challenging for a data center provider to maximize its revenue by intelligently scheduling tasks in software-defined networking (SDN)-enabled data centers. Existing SDN controllers only reduce network latency while ignoring virtual machine (VM) latency, thus may lead to revenue loss. In the context of SDN-enabled data centers, this paper presents a workload-aware revenue maximization (WARM) approach to maximize the revenue from a data center provider's perspective. The core idea is to jointly consider the optimal combination of VMs and routing paths for tasks of each application. Comparing with state-of-the-art methods, the experimental results show that WARM yields the best schedules that not only increase the revenue but also reduce the round-trip time of tasks of all applications.
Routing, Control systems, Resource management, Scheduling, Servers, Load management
H. Yuan, J. Bi, J. Zhang, W. Tan and K. Huang, "Workload-Aware Revenue Maximization in SDN-Enabled Data Center," 2017 IEEE 10th International Conference on Cloud Computing (CLOUD), Honolulu, Hawaii, United States, 2017, pp. 18-25.