Prefetching Using Principles of Hippocampal-Neocortical Interaction
Open Access
- 22 June 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 53-60
Abstract
Memory prefetching improves performance across many systems layers. However, achieving high prefetch accuracy with low overhead is challenging, as memory hierarchies and application memory access patterns become more complicated. Furthermore, a prefetcher's ability to adapt to new access patterns as they emerge is becoming more crucial than ever. Recent work has demonstrated the use of deep learning techniques to improve prefetching accuracy, albeit with impractical compute and storage overheads. This paper suggests taking inspiration from the learning mechanisms and memory architecture of the human brain---specifically, the hippocampus and neocortex---to build resource-efficient, accurate, and adaptable prefetchers.Keywords
Funding Information
- NSF (2118851, 2040682, 2112562)
This publication has 26 references indexed in Scilit:
- Contrastive Hebbian learning with random feedback weightsNeural Networks, 2019
- The Case for Learned Index Structures2018
- The storage and recall of memories in the hippocampo-cortical systemCell and tissue research, 2017
- HeteroOSACM SIGARCH Computer Architecture News, 2017
- Perceptron learning for reuse prediction2016
- What Learning Systems do Intelligent Agents Need? Complementary Learning Systems Theory UpdatedTrends in Cognitive Sciences, 2016
- Deep Residual Learning for Image Recognition2016
- Pattern separation, completion, and categorisation in the hippocampus and neocortexNeurobiology of Learning and Memory, 2016
- Workload analysis of a large-scale key-value store2012
- Long Short-Term MemoryNeural Computation, 1997