LLCG: A High Performance Implement for Multi-tenant Data Placement

How to optimally place the tenant replication data to retain the load-balance and reduce cost of communication and distributed transaction, it is an important issue to achieve the high performance and availability of multi-tenant data, there are plenty of issues need to be solved. This paper proposes the Multi-Objective Genetic Algorithm. It uses a rank-based fitness assignment method for MOGAs to placement and adjustment the multi-tenant data called LLCG. Then we validate the effectiveness and performance of our algorithm compared with LRCG and LLC in simulation experiment.