Title PMDB : Partition-based Multi-instance Database System for Multicore Platforms

The continued evolution of modern hardware has brought seve ral n w challenges to database management systems (DBMSs). Multicore CPUs are now mainstream, while t e future lies in massively parallel computing performed on many-core processors. However, because th ey were developed originally for single-core processors, DBMSs cannot take full advantage of the paralle l computing that uses so many cores. Several components in the traditional database engines become new b ottlenecks on multicore platforms. In this paper, we analyze the bottlenecks in existing database engine s o a modern multicore platform using the mixed workload of the TPC-W benchmark and describe strategies for higher scalability and throughput for existing DBMSs on multicore platforms. First, we show how to overc ome the limitations of the database engine by introducing a partition-based multi-instance database system on a single multicore platform without any modification of existing DBMSs. Second, we analyze the possi bility of further improving the performance by optimizing the cache performance of concurrent queries. Implemented by middleware, our proposed PMDB can avoid the challenging work of modifying existing data base engines. Performance evaluation using the TPC-W benchmark revealed that our proposal can achieve a t most 2.5 times higher throughput than the existing engine of PostgreSQL.