Heracles
David Lo, Liqun Cheng, Rama Govindaraju et al. · Proceedings - International Symposium on Computer Architecture/Proceedings · 2015
User-facing, latency-sensitive services, such as websearch, underutilize their computing resources during daily periods of low traffic. Reusing those resources for other tasks is rarely done in production services since the contention for shared resources can cause latency spikes that violate the service-level objectives of latency-sensitive tasks. The resulting under-utilization hurts both the affordability and energy-efficiency of large-scale datacenters. With technology scaling slowing down, it becomes important to address this opportunity.
Cited by 466 works. doi:10.1145/2749469.2749475 Open-access PDF
Authors
References
- The Case for Energy-Proportional Computing
- Building high-level features using large scale unsupervised learning
- A scalable, commodity data center network architecture
- Building high-level features using large scale unsupervised learning
- A scalable, commodity data center network architecture
- The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines
- Data center TCP (DCTCP)
- The tail at scale
- PowerNap
- Dark silicon and the end of multicore scaling