Fast Modeling L2 Cache Reuse Distance Histograms Using Combined Locality Information from Software Traces

TL;DR

This paper introduces a fast, accurate model for predicting L2 cache reuse distance histograms using software trace locality information, significantly speeding up cache performance evaluation during processor design.

Contribution

It presents a novel L2 cache reuse distance histogram model utilizing L1 reuse data and new metrics, enabling rapid and accurate cache behavior prediction across configurations.

Findings

Average absolute error less than 5% in miss rate predictions

Evaluation time reduced by nearly 30 times for multiple cache configurations

Model validated on SPEC CPU benchmarks with high accuracy

Abstract

To mitigate the performance gap between CPU and the main memory, multi-level cache architectures are widely used in modern processors. Therefore, modeling the behaviors of the downstream caches becomes a critical part of the processor performance evaluation in the early stage of Design Space Exploration (DSE). In this paper, we propose a fast and accurate L2 cache reuse distance histogram model, which can be used to predict the behaviors of the multi-level cache architectures where the L1 cache uses the LRU replacement policy and the L2 cache uses LRU/Random replacement policies. We use the profiled L1 reuse distance histogram and two newly proposed metrics, namely the RST table and the Hit-RDH, that describing more detailed information of the software traces as the inputs. For a given L1 cache configuration, the profiling results can be reused for different configurations of the L2 cache. The output of our model is the L2 cache reuse distance histogram, based on which the L2 cache miss rates can be evaluated. We compare the L2 cache miss rates with the results from gem5 cycle-accurate simulations of 15 benchmarks chosen from SPEC CPU 2006 and 9 benchmarks from SPEC CPU 2017. The average absolute error is less than 5%, while the evaluation time for each L2 configuration can be sped up almost 30X for four L2 cache candidates.