Proposed Congestion Control Method for Cloud Computing Environments

摘要

As cloud computing services rapidly expand their customer base, it has become important to share cloud resources, so as to provide them economically. In cloud computing services, multiple types of resources, such as processing ability, bandwidth and storage, need to be allocated simultaneously. If there is a surge of requests, a competition will arise between these requests for the use of cloud resources. This leads to the disruption of the service and it is necessary to consider a measure to avoid or relieve congestion of cloud computing environments. This paper proposes a new congestion control method for cloud computing environments which reduces the size of required resource for congested resource type instead of restricting all service requests as in the existing networks. Next, this paper proposes the user service specifications for the proposed congestion control method, and clarifies the algorithm to decide the optimal size of required resource to be reduced, based on the load offered to the system. It is demonstrated by simulation evaluations that the proposed method can handle more requests compared with the conventional methods and relieve the congestion. Then, this paper proposes to enhance the proposed method, so as to enable the fair resource allocation among users in congested situation.