C-Hint: An effective and reliable cache management for RDMA-accelerated key-value stores

Abstract

Recently, many in-memory key-value stores have started using a High-Performance network protocol, Remote Direct Memory Access (RDMA), to provision ultra-low latency access services. Among various solutions, previous studies have recognized that leveraging RDMA Read to optimize GET operations and continuing using message passing for other requests can offer tremendous performance improvement while avoiding read-write races. However, although such a design can utilize the power of RDMA when there is sufficient memory space, it has also raised new challenges on the cache management that do not exist in traditional key-value stores. First, RDMA Read deprives servers of the awareness of the read operations. Therefore, how to track popular items and make replacement decisions at the server side becomes a critical issue. Second, without the access knowledge from the clients, new approaches are needed for servers to efficiently and reliably reclaim the resources. Lastly, the remote pointers hold by clients to conduct RDMA are highly susceptible to the evictions made by remote servers. Thus, any replacement algorithm that solely considers the server-side hit ratio is insufficient and can cause severe underutilization of RDMA. In this work, we present C-Hint, an efficient and reliable cache management system that is designed to address the above three challenges. Its basic mechanism relies on a holistic design of the servers and the clients to orchestrate the access history information. Specifically, it consists of several techniques, including lease-based key-value management, popularity differentiated lease assignment, as well as several optimizations to achieve efficient and reliable cache management. We have implemented and integrated C-Hint into an in-house memory-resident key-value store, named HydraDB, and systematically evaluated its performance on an InfiniBand cluster with different workloads. Our experiment results demonstrate that C-Hint can efficiently outperform several other alternate solutions and deliver both high hit rate and low latency performance.