Tight Cache Contention Analysis for WCET Estimation on Multicore Systems

TL;DR

This paper introduces a precise cache contention analysis method for multicore WCET estimation, significantly reducing overestimations by considering actual cache states and access patterns, leading to more accurate timing bounds.

Contribution

It presents a novel fine-grained inter-core cache contention analysis that improves WCET estimation accuracy by modeling cache behavior more precisely.

Findings

Reduces inter-core cache interference estimates by 52.31%.

Decreases WCET estimations by 8.94% on average.

Maintains low computational overhead.

Abstract

WCET (Worst-Case Execution Time) estimation on multicore architecture is particularly challenging mainly due to the complex accesses over cache shared by multiple cores. Existing analysis identifies possible contentions between parallel tasks by leveraging the partial order of the tasks or their program regions. Unfortunately, they overestimate the number of cache misses caused by a remote block access without considering the actual cache state and the number of accesses. This paper reports a new analysis for inter-core cache contention. Based on the order of program regions in a task, we first identify memory references that could be affected if a remote access occurs in a region. Afterwards, a fine-grained contention analysis is constructed that computes the number of cache misses based on the access quantity of local and remote blocks. We demonstrate that the overall inter-core cache interference of a task can be obtained via dynamic programming. Experiments show that compared to existing methods, the proposed analysis reduces inter-core cache interference and WCET estimations by 52.31% and 8.94% on average, without significantly increasing computation overhead.