Last updated on 11/04/2019
Huang, Z. , Joao, J. A. , Rico, A. , Hilton, A. D. , & Lee, B. C. . (2019). DynaSprint: Microarchitectural Sprints with Dynamic Utility and Thermal Management. In In Proceedings of the 52nd IEEE/ACM International Symposium on Microarchitecture (pp. 426-439). Retrieved from https://dl.acm.org/citation.cfm?id=3358301
Sprinting is a class of mechanisms that provides a short but significant performance boost while temporarily exceeding the thermal design point. We propose DynaSprint, a software runtime that manages sprints by dynamically predicting utility and modeling thermal headroom. Moreover, we propose a new sprint mechanism for caches, increasing capacity briefly for enhanced performance. For a system that extends last-level cache capacity from 2MB to 4MB per core and can absorb 10J of heat, DynaSprint-guided cache sprints improve performance by 17% on average and by up to 40% over a non-sprinting system. These performance outcomes, within 95% of an oracular policy, are possible because DynaSprint accurately predicts phase behavior and sprint utility.