Building blocks for persistent memory

Abstract

I/O latency and throughput are two of the major performance bottlenecks for disk-based database systems. Persistent memory (PMem) technologies, like Intel’s Optane DC persistent memory modules, promise to bridge the gap between NAND-based flash (SSD) and DRAM, and thus eliminate the I/O bottleneck. In this paper, we provide the first comprehensive performance evaluation of PMem on real hardware in terms of bandwidth and latency. Based on the results, we develop guidelines for efficient PMem usage and four optimized low-level building blocks for PMem applications: log writing, block flushing, in-place updates, and coroutines for write latency hiding.

ESH: Design and Implementation of an Optimal Hashing Scheme for Persistent Memory

“…In the case of search operations, Dash and ESH show near-linear performance, while CCEH lags behind due to the impact of locking and cache-line-level hashing, which is affected by the size of the hash table and requires a larger number of persistent memory writes. Although all three schemes employ similar extendible hashing techniques, the differences lie in the design of the lock strips [55], which allows locking to fit into the CPU cache, and the impact of the building blocks of persistent memory on log writing and block flushing [56], which affects write bandwidth. ESH demonstrates higher performance in insertion, search, and delete operations and scales effectively with varying numbers of threads.…”