Scalable persistent memory (PM), represented by the Intel Optane PMem, has opened new opportunities for building indexes that operate and persist data directly on the memory bus, potentially enabling instant recovery, low latency, and high throughput. Since its first release in 2019, newer PM indexes have emerged, but it is unclear how they compare to each other and to previous proposals in terms of both design and performance, and what future challenges remain. To address these issues, we comprehensively analyzed and experimentally evaluated state-of-the-art PM range indexes under a unified benchmark. The results suggest newer designs inherited past techniques, but some do not necessarily outperform pre-Optane era proposals and can be improved. Functionality wise, most indexes still lack good support for variable-length keys and handling NUMA effect. Based on our findings, we gave a list of observations, insights, and future directions with consideration for current changes in related industry.
Copyright is held by the author(s).
This thesis may be printed or downloaded for non-commercial research and scholarly purposes.
Supervisor or Senior Supervisor
Thesis advisor: Wang, Tianzheng
Member of collection