Advanced Cache Techniques for SLA-Driven Multi-Tenant Application on PaaS

Abstract

Multi-tenant application is one of the main characteristics of cloud computing. Today, most of the application uses cache service for getting faster access and low response time. Currently in multi-tenant cloud applications data are often evicted mistakenly by cache service, which is managed by existing algorithms such as LRU. Also, security mechanisms are implemented to avoid data breach when data are accessed improperly by another tenant. SLA Driven cache optimization Approach for multi-tenant application is built on PaaS. It helps to improve the cache performance and cost effectiveness of tenants. This tries to improve the cache utilization by avoiding faulty evictions and unnecessary storage retrievals. It considers both tenant profile and data profile in addition to LRU for weight the evicted data and determines re-cache mechanism. Memcache is currently used to cache data explicitly. It provides only average response time to users. In the proposed work, HashMap is introduced as a new form of internal cache on PaaS for faster access. Tenants can access this application through web service by using PC, Laptop, Mobile etc. The test results shows that the proposed method provides better performance and faster access to multi-tenant users.