FORD: Fast One-sided RDMA-based Distributed Transactions for Disaggregated Persistent Memory

Persistent memory (PM) disaggregation improves the resource utilization and failure isolation to build a scalable and cost-effective remote memory pool. However, due to offering limited computing power and overlooking the bandwidth and persistence properties of real PMs, existing distributed transaction schemes, which are designed for legacy DRAM-based monolithic servers, fail to efficiently work in the disaggregated PM architecture. In this paper, we propose FORD, a Fast One-sided RDMA-based Distributed transaction system. FORD thoroughly leverages one-sided RDMA to handle transactions for bypassing the remote CPU in PM pool. To reduce the round trips, FORD batches the read and lock operations into one request to eliminate extra locking and validations. To accelerate the transaction commit, FORD updates all the remote replicas in a single round trip with parallel undo logging and data visibility control. Moreover, considering the limited PM bandwidth, FORD enables the backup replicas to be read to alleviate the load on the primary replicas, thus improving the throughput. To efficiently guarantee the remote data persistency in the PM pool, FORD selectively flushes data to the backup replicas to mitigate the network overheads. Experimental results demonstrate that FORD improves the transaction throughput by up to 2.3x and reduces the latency by up to 74.3% compared with the state-of-the-art systems.