ParRP: Enabling Space Isolation in Caches with Shared Data

• Published in: Proceedings / IEEE Real-Time and Embedded Technology and Applications Symposium pp. 82 - 94
• Main Authors: Wang, Xinzhe, Pellizzoni, Rodolfo, Patel, Hiren
• Format: Conference Proceeding
• Language: English
• Published: IEEE 06.05.2025
• Subjects: cache partitioning; Costs; Full stack; Hardware; Interference; Memory; Multicore processing; predictable architecture; Real-time systems; resource isolation; shared last-level cache
• ISSN: 2642-7346
• DOI: 10.1109/RTAS65571.2025.00020

This work presents an approach to isolate cache space while supporting shared data. Enabling shared data caching is challenging because it causes interference making it unsuitable for real-time multicores. Unlike prior works, we aim to introduce isolation, but our approach enables caching of shared data, and promotes isolated cache analysis for individual cores. The crux behind our approach is that shared data isolation can be achieved by partitioning the replacement information instead of the cache's data storage. Consequently, this work introduces ParRP, a novel hardware cache partitioning scheme for realtime cache-coherent multicores. Our evaluation using the gem5 simulator shows that, by providing isolation for shared data, the worst-case execution time of multi-threaded tasks can be lower by 2.4× at the cost of a 16.5% decrease in average-case performance.