Optimizing Data Structures for Persistent Memory

• Grant number: 23/04969-8
• Support Opportunities: Scholarships in Brazil - Master
• Start date: May 01, 2023
• End date: April 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Alexandro José Baldassin
• Grantee: Lucas Bastelli Spagnol
• Host Institution: Instituto de Geociências e Ciências Exatas (IGCE). Universidade Estadual Paulista (UNESP). Campus de Rio Claro. Rio Claro , SP, Brazil
• Associated research grant: 18/15519-5 - Performance optimizations for multicore architectures, AP.JP2

Abstract

Persistent Memory (PM) is an emerging memory technology aimed to eliminate the gap between main memory and stable storage. Nevertheless, today's programs will not readily benefit from PM because crash failures may render the program in an unrecoverable and inconsistent state. In particular, it is of great interest to investigate how common data structures can be modified to be used transparently with PM in an efficient way. This problem is complicated even further given that current platform employ both volatile (i.e., DRAM) and PM at the same memory hierarchy level. Therefore, devising new techniques that use both volatile and non-volatile memory is a challenge and many researchers have come up with different solutions. This MSc. project seeks to advance the state-of-the-art of data structures design for PM. In a previous work, we have conducted an analysis of the performance of typical data structures (e.g., linked list, hashmap) on PM and noticed that they fell short of achieving optimum performance. In this work weseek to improve the performance by devising new hybrid solutions to the problem.