Providing Fairness in Heterogeneous Multicores with a Predictive, Adaptive Scheduler
| Autor: | Henry Hoffmann, Saeid Barati |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
010302 applied physics
Multi-core processor Computer science Distributed computing Processor scheduling Memory bandwidth Workload 02 engineering and technology 01 natural sciences 020202 computer hardware & architecture Scheduling (computing) Instruction set 0103 physical sciences 0202 electrical engineering electronic engineering information engineering |
| Zdroj: | IPDPS Workshops |
| Popis: | Multicore applications contend for resources -- especially memory bandwidth -- reducing both quality-of-service and overall system performance. Contention-awareschedulers have been proposed to provide fairness and predictable behavior through thread-level scheduling. Prior approaches have two drawbacks, however. First, many introduce overhead that reduces overall performance. Second, the emergence of heterogeneous multicores has made handlingcontention and providing fairness much more difficult as thescheduler must now account for both application interferenceand the performance effects of different core types. This paper proposes augmenting existing contention-awareapproaches with predictive and adaptive components to providefair memory access and performance improvements on heterogeneous multicores. The predictive component's closed-loopapproach anticipates how different processes will perform withdifferent core types, while the adaptive component dynamicallytunes key scheduling parameters to the current workload. Weimplement and evaluate this approach on a real Linux/x86system with a variety of memory and compute intensivebenchmarks. We find that adding prediction improves fairnessand performance by 38% and 4% (respectively) compared to aprior state-of-the-art contention-aware approach. The additionof adaptation allows users to select for fairness or performanceoptimization, providing an additional 24% improvement infairness or a 9% improvement in performance beyond the predictive approach. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|