Design And Implementation Of High-Performance Server Architecture Of SAAS

With the rapid development of Internet technology and changes in the software business model, making the method of the software design, development, delivery changed. Currently, reliable bandwidth and the emergence of service-oriented architecture makes the software as a service model delivered to the customer, that the SaaS model has become mainstream. Customers do not have to spend a lot of manpower and resources to develop their own software, and they only need to pay to software services providers and get software service.So,this model get more and more popular.However, with the increasing number of tenants, resulting in the number of requests to the server and load suddenly increases, which increase the software service provider’s server infrastructure requirements. How to ensure the stability and performance of the entire structure as well as smaller response delay, providing high-quality, uninterrupted software services when a large number of concurrent users accessing the system becomes the problem which software service providers must be resolved. Obviously a single server architecture can not meet the demand in terms of reliability, performance. The cluster technology which widely used in recent years has a very good performance, reliability, scalability.Based on the research of LVS cluster technology, we propose a high-performance server architecture of SaaS and then implement this architecture,By testing this architecture,we prove it has good performance, reliability and scalability. Secondly, we analyzes the deficiencies of traditional load balancing scheduling algorithm, and propose an improved algorithm based on multi-tenant SaaS model and validate this algorithm by simulation tests. Finally, analyze the caching algorithms and propose improved caching algorithm based on multi-tenant SaaS model and do simulation tests to verify its better.