Towards Multitenancy for IO-Bound OLAP Workloads
Consolidation of multiple databases on the same server allows service providers to save significant resources because many production database servers are often under-utilized. The authors consider the problem of minimizing the number of servers needed to host a set of tenants, while satisfying the Service Level Agreement (SLA) on the throughput of each tenant. Recent research investigates this problem under the assumption that the working sets of tenants are kept in main memory (e.g., OLTP workloads, or in-memory OLAP workloads), thus the buffer size of each tenant is dictated by the working set size of that tenant. In this paper, they instead investigate the problem when the throughput SLAs of tenants are low enough for queries to be answered from disk.