Performance Guarantee Mechanism for Multi-Tenancy SaaS Service Based on Kalman Filtering

Abstract

Abstract This paper proposes a special System Architecture for Multi-tenancy SaaS Service (SAMSS), which studies the performance security issues at the business logic layer and data processing layer respectively. The Kalman filtering Admission Control algorithm (KAC) and the Greedy Copy Management algorithm (GCM) are proposed. At the business logic layer, Kalman filtering admission control algorithm is presented. It uses a Kalman filter to conduct the dynamic evaluation for the CPU resource for multi-tenancy SaaS service and reduces the unnecessary performance expenses caused by direct measurement of CPU resources. At the data processing layer, the Greedy Copy Management algorithm (GCM) is presented. It changes the copy placement as a K-partitioning set partitioning problem and adopts a greedy strategy to reduce the number of times for creating a data copy. Finally, the experimental analysis and results prove the feasibility and efficiency of the algorithms proposed.