SWORD: scalable workload-aware data placement for transactional workloads

In this paper, we address the problem of transparently scaling out transactional (OLTP) workloads on relational databases, to support database-as-a-service in cloud computing environment. The primary challenges in supporting such workloads include choosing how to partition the data across a large number of machines, minimizing the number of distributed transactions, providing high data availability, and tolerating failures gracefully. Capturing and modeling the transactional workload over a period of time, and then exploiting that information for data placement and replication has been shown to provide significant benefits in performance, both in terms of transaction latencies and overall throughput. However, such workload-aware data placement approaches can incur very high overheads, and further, may perform worse than naive approaches if the workload changes. In this work, we propose SWORD, a <u>s</u>calable <u>wor</u>kload-aware <u>d</u>ata partitioning and placement approach for OLTP workloads, that incorporates a suite of novel techniques to significantly reduce the overheads incurred both during the initial placement, and during query execution at runtime. We model the workload as a hypergraph over the data items, and propose using a hypergraph compression technique to reduce the overheads of partitioning. To deal with workload changes, we propose an incremental data repartitioning technique that modifies data placement in small steps without resorting to complete workload repartitioning. We have built a workload-aware active replication mechanism in SWORD to increase availability and enable load balancing. We propose the use of fine-grained quorums defined at the level of groups of tuples to control the cost of distributed updates, improve throughput, and provide adaptability to different workloads. To our knowledge, SWORD is the first system that uses fine-grained quorums in this context. The results of our experimental evaluation on SWORD deployed on an Amazon EC2 cluster show that our techniques result in orders-of-magnitude reductions in the partitioning and book-keeping overheads, and improve tolerance to failures and workload changes; we also show that choosing quorums based on the query access patterns enables us to better handle query workloads with different read and write access patterns.