On Usage of Non-Volatile Memory as Primary Storage for Database Management Systems

TL;DR

This paper investigates using non-volatile memory as primary storage in database systems, modifies PostgreSQL's storage engine for NVM, and introduces a prefetching library to improve performance, achieving significant query time reductions.

Contribution

It presents the first detailed modifications of a relational DBMS to leverage NVM and introduces a prefetching library to mitigate latency issues, enhancing performance.

Findings

Modified storage engine reduces query time by up to 45% compared to disk.

Prefetching library further improves query time by up to 54%.

Data access efficiency is improved, but latency issues remain.

Abstract

This paper explores the implications of employing non-volatile memory (NVM) as primary storage for a data base management system (DBMS). We investigate the modifications necessary to be applied on top of a traditional relational DBMS to take advantage of NVM features. As a case study, we modify the storage engine (SE) of PostgreSQL enabling efficient use of NVM hardware. We detail the necessary changes and challenges such modifications entail and evaluate them using a comprehensive emulation platform. Results indicate that our modified SE reduces query execution time by up to 45% and 13% when compared to disk and NVM storage, with average reductions of 19% and 4%, respectively. Detailed analysis of these results shows that while our modified SE is able to access data more efficiently, data is not close to the processing units when needed for processing, incurring long latency misses that hinder the performance. To solve this, we develop a general purpose library that employs helper threads to prefetch data from NVM hardware via a simple API. Our library further improves query execution time for our modified SE when compared to disk and NVM storage by up to 54% and 17%, with average reductions of 23% and 8%, respectively.