Bandits meet Computer Architecture: Designing a Smartly-allocated Cache

TL;DR

This paper introduces a bandit-based approach for dynamic cache allocation in embedded systems, enabling online learning and optimization of resource distribution to improve system performance.

Contribution

It presents novel online and offline algorithms for resource allocation using multi-armed bandits, tailored for embedded system cache management.

Findings

Algorithms outperform baseline methods in synthetic tests

Effective online learning of cache impact on threads

Improved system performance through adaptive resource allocation

Abstract

In many embedded systems, such as imaging sys- tems, the system has a single designated purpose, and same threads are executed repeatedly. Profiling thread behavior, allows the system to allocate each thread its resources in a way that improves overall system performance. We study an online resource al- locationproblem,wherearesourcemanagersimulta- neously allocates resources (exploration), learns the impact on the different consumers (learning) and im- proves allocation towards optimal performance (ex- ploitation). We build on the rich framework of multi- armed bandits and present online and offline algo- rithms. Through extensive experiments with both synthetic data and real-world cache allocation to threads we show the merits and properties of our al- gorithms