CIDER: Boosting Memory-Disaggregated Key-Value Stores with Pessimistic Synchronization

TL;DR

CIDER introduces a pessimistic synchronization framework with global write-combining and contention-aware schemes to significantly reduce redundant I/O and boost throughput in memory-disaggregated key-value stores.

Contribution

It proposes a novel pessimistic synchronization approach with specific optimizations for memory-disaggregated KV stores, addressing I/O redundancy issues.

Findings

Up to 6.6x throughput improvement on YCSB benchmark.

Effective reduction of cross-node redundant I/Os.

Enhanced performance under low contention scenarios.

Abstract

Memory-disaggregated key-value (KV) stores suffer from a severe performance bottleneck due to their I/O redundancy issues. A huge amount of redundant I/Os are generated when synchronizing concurrent data accesses, making the limited network between the compute and memory pools of DM a performance bottleneck. We identify the root cause for the redundant I/O lies in the mismatch between the optimistic synchronization of existing memory-disaggregated KV stores and the highly concurrent workloads on DM. In this paper, we propose to boost memory-disaggregated KV stores with pessimistic synchronization. We propose CIDER, a compute-side I/O optimization framework, to verify our idea. CIDER adopts a global write-combining technique to further reduce cross-node redundant I/Os. A contention-aware synchronization scheme is designed to improve the performance of pessimistic synchronization under low contention scenarios. Experimental results show that CIDER effectively improves the throughput of state-of-the-art memory-disaggregated KV stores by up to $6.6\times$ under the YCSB benchmark.