Cache Bypassing for Machine Learning Algorithms

TL;DR

This paper introduces a predictive cache bypassing model for GPU-based machine learning algorithms, significantly improving cache efficiency and accuracy in data access pattern analysis.

Contribution

It proposes a novel predictive model that determines cache storage decisions for machine learning workloads on GPUs, enhancing performance and cache utilization.

Findings

Model achieves around 90% accuracy on various datasets.

KNN performs best among tested models.

Splitting addresses into chunks improves neural network accuracy to 99.9%.

Abstract

Graphics Processing Units (GPUs) were once used solely for graphical computation tasks but with the increase in the use of machine learning applications, the use of GPUs to perform general-purpose computing has increased in the last few years. GPUs employ a massive amount of threads, that in turn achieve a high amount of parallelism, to perform tasks. Though GPUs have a high amount of computation power, they face the problem of cache contention due to the SIMT model that they use. A solution to this problem is called "cache bypassing". This paper presents a predictive model that analyzes the access patterns of various machine learning algorithms and determines whether certain data should be stored in the cache or not. It presents insights on how well each model performs on different datasets and also shows how minimizing the size of each model will affect its performance The performance of most of the models were found to be around 90% with KNN performing the best but not with the smallest size. We further increase the features by splitting the addresses into chunks of 4 bytes. We observe that this increased the performance of the neural network substantially and increased the accuracy to 99.9% with three neurons.