An Efficient and Fair Multi-Resource Allocation Mechanism for Heterogeneous Servers
Abstract
Introduction
An Efficient and Fair Multi-Resource Allocation Mechanism for Heterogeneous Servers As it provides a cost-effective alternative to proprietary high-performance computing systems, cloud computing has become increasingly popular. As the workload to data centers housing cloud computing platforms is growing intensively, it has become increasingly important to develop an efficient and fair allocation mechanism that guarantees quality-ofservice for different workloads. In such a shared computing system, efficient and fair allocation of resources is particularly challenging due to the presence of multiple types of resources, diversity in the needs of the workloads for these resources, heterogeneity in the resource capacities of servers.
System Configuration
Platform : cloud computing
Conclusion
In the presence of placement constraints, we studied efficient and fair allocation of multiple types of resources in an environment of heterogeneous servers. In existing multi-resource allocation mechanisms, DRF and its follow-up work, we identified potential limitations when used in such environments. They may not succeed in satisfying all the essential fairness-related properties on certain occasions, may not be readily implementable in a distributed fashion, and may lead to inefficient use of resources. We proposed a new server-based approach to efficiently allocate resources while capturing heterogeneity of the server.
