Professor Janki Bhimani is leading a new NSF Award to apply Machine Learning techniques to design tiered memory management systems for large-scale memory-intensive applications. In the project, she will explore the power of machine learning (ML) as well as its limits and overheads to develop a more general, adaptable solution for improving tiered memory management systems for large-scale memory-intensive applications. Many aspects involved in these memory systems such as scanning, migration, allocation, parameter tuning, and task scheduling to optimize the performance, quality of service, and resource utilization will be addressed in the research. The project goal is to design a better-tiered memory system for rapidly changing workloads.
This is a fantastic project, which is highly relevant to today’s technology development.
NSF will fund the project which totals $514,704 over three years. A full project abstract can be found on the NSF award page at NSF Award Search: Award # 2323100 – CSR: Small: Learning and Management in Tiered Memory Systems.