Data Center Sprinting: Enabling Computational Sprinting at the Data Center Level

Zheng, Wenli; Wang, Xiaorui

doi:10.1109/icdcs.2015.26

Cited by 29 publications

(9 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Computational sprinting enables bursts of peak processing in systems limited by dark silicon. Sprinting mechanisms exist nowadays at all system levels, from transistors and processors [35], to computer systems [26], and datacenters [44]. Sprinting policies decide when and what to sprint, e.g., phases within job executions [43] and particular queries [25].…”

Section: Related Workmentioning

confidence: 99%

Differential Approximation and Sprinting for Multi-Priority Big Data Engines

Birke

Rocha

Pérez

et al. 2019

Proceedings of the 20th International Middleware Conference

View full text Add to dashboard Cite

Today's big data clusters based on the MapReduce paradigm are capable of executing analysis jobs with multiple priorities, providing differential latency guarantees. Traces from production systems show that the latency advantage of highpriority jobs comes at the cost of severe latency degradation of low-priority jobs as well as daunting resource waste caused by repetitive eviction and re-execution of low-priority jobs. We advocate a new resource management design that exploits the idea of differential approximation and sprinting. The unique combination of approximation and sprinting avoids the eviction of low-priority jobs and its consequent latency degradation and resource waste. To this end, we designed, implemented and evaluated DiAS, an extension of the Spark processing engine to support deflate jobs by dropping tasks and to sprint jobs. Our experiments on scenarios with two and three priority classes indicate that DiAS achieves up to 90% and 60% latency reduction for low-and high-priority jobs, respectively. DiAS not only eliminates resource waste but also (surprisingly) lowers energy consumption up to 30% at only a marginal accuracy loss for low-priority jobs.

show abstract

Section: Related Workmentioning

confidence: 99%

Differential Approximation and Sprinting for Multi-Priority Big Data Engines

Birke

Rocha

Pérez

et al. 2019

Proceedings of the 20th International Middleware Conference

View full text Add to dashboard Cite

show abstract

“…We consider circuit breakers that can be overloaded to 125-175% of rated current for a 150s sprint. 18,21 Uninterruptible power supplies. When the breaker trips and resets, power distribution switches from the branch circuit to the uninterruptible power supply (UPS).…”

Section: The Sprinting Gamementioning

confidence: 99%

“…It can accelerate computation for complex tasks or accommodate transient activity spikes. 16,21 The system architecture determines sprint duration and frequency. Sprinting multiprocessors generate extra heat, absorbed by thermal packages and phase change materials (PCMs), 14,16 and require time to release this heat between sprints.…”

Section: Introductionmentioning

confidence: 99%

Distributed strategies for computational sprints

2019

View full text Add to dashboard Cite

Computational sprinting is a class of mechanisms that boost performance but dissipate additional power. We describe a sprinting architecture in which many, independent chip multiprocessors share a power supply and sprints are constrained by the chips' thermal limits and the rack's power limits. Moreover, we present the computational sprinting game, a multi-agent perspective on managing sprints. Strategic agents decide whether to sprint based on application phases and system conditions. The game produces an equilibrium that improves task throughput for data analytics workloads by 4-6× over prior greedy heuristics and performs within 90% of an upper bound on throughput from a globally optimized policy.

show abstract

“…In each round, 1000 agents compete for 100 power boosts. These parameters reflect modern power supplies [13,70] and our workloads' typical power demands.…”

Section: Game Dynamicsmentioning

confidence: 99%

Managing Heterogeneous Datacenters with Tokens

Zahedi

Fan

Lee

2018

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Ensuring fairness in a system with scarce, preferred resources requires time sharing. We consider a heterogeneous system with a few "big" and many "small" processors. We allocate heterogeneous processors using a novel token mechanism, which frames the allocation problem as a repeated game. At each round, users request big processors and spend a token if their request is granted. We analyze the game and optimize users' strategies to produce an equilibrium. In equilibrium, allocations balance performance and fairness. Our mechanism outperforms classical, fair mechanisms by 1.7×, on average, in performance gains, and is competitive with a performance maximizing mechanism. CCS Concepts: • Software and its engineering → Power management; • Information systems → Data centers; • Social and professional topics → Pricing and resource allocation; • Theory of computation → Algorithmic mechanism design; • Computing methodologies → Modeling methodologies;

show abstract

Data Center Sprinting: Enabling Computational Sprinting at the Data Center Level

Cited by 29 publications

References 25 publications

Differential Approximation and Sprinting for Multi-Priority Big Data Engines

Differential Approximation and Sprinting for Multi-Priority Big Data Engines

Distributed strategies for computational sprints

Managing Heterogeneous Datacenters with Tokens

Contact Info

Product

Resources

About