Energy-aware SRPT server with batch arrivals: Analysis and optimization

Gebrehiwot, Misikir Eyob; Aalto, Samuli; Lassila, Pasi

doi:10.1016/j.peva.2017.07.003

Cited by 10 publications

(16 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Under more general assumptions with general service and setup times, they computed the two mean values [R] and [E] by means of some queueing insensitivity, and then discussed the optimality of the ERP and ERWS. Further, Gebrehiwot et al [16,17] generalized the FCFS queueing results of the data center with multiple intermediate sleep states to the processor-sharing discipline and the shortest remaining processing time (SRPT) discipline, respectively. Different from the ERP and ERWS, Mitrani [35,36] considered a data center of N identical servers whose first part contains m servers.…”

Section: Introductionmentioning

confidence: 99%

Optimal energy-efficient policies for data centers through sensitivity-based optimization

Xia

2019

Discrete Event Dyn Syst

View full text Add to dashboard Cite

In this paper, we propose a novel dynamic decision method by applying the sensitivity-based optimization theory to find the optimal energy-efficient policy of a data center with two groups of heterogeneous servers. Servers in Group 1 always work at high energy consumption, while servers in Group 2 may either work at high energy consumption or sleep at low energy consumption. An energy-efficient control policy determines the switch between work and sleep states of servers in Group 2 in a dynamic way. Since servers in Group 1 are always working with high priority to jobs, a transfer rule is proposed to migrate the jobs in Group 2 to idle servers in Group 1. To find the optimal energy-efficient policy, we set up a policy-based Poisson equation, and provide explicit expressions for its unique solution of performance potentials by means of the RG-factorization. Based on this, we characterize monotonicity and optimality of the long-run average profit with respect to the policies under different service prices. We prove that the bang-bang control is always optimal for this optimization problem, i.e., we should either keep all servers sleep or turn on the servers such that the number of working servers equals that of waiting jobs in Group 2. As an easy adoption of policy forms, we further study the threshold-type policy and obtain a necessary condition of the optimal threshold policy. We hope the methodology and results derived in this paper can shed light to the study of more general energy-efficient data centers.

show abstract

Section: Introductionmentioning

confidence: 99%

Optimal energy-efficient policies for data centers through sensitivity-based optimization

Xia

2019

Discrete Event Dyn Syst

View full text Add to dashboard Cite

show abstract

“…Remark: Note that the regenerative cycle definition is identical with our earlier papers [7,6,8]. However, without any major modifications to the analysis the regeneration point could also be selected as the point where the queue empties.…”

Section: We Denote By E[t ] the Overall Mean Delay Of The Jobs And Bymentioning

confidence: 99%

“…Energy-aware queueing models have been recently developed in order to study the performance-energy tradeoff inherent in modern data servers supporting sleep states. The considered queueing models are typically variations of the single-server M/G/1 queue with a setup delay, where the setup delay reflects the delay penalty from waking up a server once it has been put to sleep to save energy, see [3,6,7,8,10]. Also, multisever variants of the models have been studied, see [2,3,4,11].…”

Section: Introductionmentioning

confidence: 99%

“…The model has been analyzed for different scheduling disciplines, including FIFO [7,10], PS [6] and SRPT [8]. In this paper, we consider the same model but assume the Least Attained Service (LAS) scheduling discipline, sometimes also referred to as Foreground Background (FB), which always serves the job with the least amount of attained service.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On energy-aware M/G/1-LAS queue with batch arrivals

Lassila

Aalto

2019

Operations Research Letters

Self Cite

View full text Add to dashboard Cite

“…In [7], such servers were called InstantOff servers, while the ordinary ones were referred to as NeverOff servers. In [23,10,9,11], it was shown, under various assumptions, that the optimal sleep state control policy for a single server is either InstantOff or NeverOff. The FPI approach has been applied also for the dispatching problem in the context of such energy-aware InstantOff servers, see, e.g., [18,17,16,12,19,14].…”

Section: Introductionmentioning

confidence: 99%

Near-optimal dispatching policy for energy-aware server clusters

Aalto

Lassila

2019

Performance Evaluation

Self Cite

View full text Add to dashboard Cite

A server cluster can be modeled as a set of parallel queues, and the dispatcher decides to which queue the arriving jobs are routed. We consider an energy-aware dispatching system in a Markovian setting, where each server, upon becoming empty, enters a sleep mode to save energy, and to activate the server after sleep incurs an additional setup delay cost. We seek to optimize the performance-energy trade-off by applying the so-called Whittle index approach. As our main result, we derive sufficient conditions for the system parameters under which the problem is provably indexable, and also determine the corresponding Whittle index values explicitly. Our numerical experiments demonstrate that the resulting energy-aware Whittle index policy is able to perform very close to the numerically solved optimal policy and outperforms all considered reference policies.

show abstract

Energy-aware SRPT server with batch arrivals: Analysis and optimization

Cited by 10 publications

References 15 publications

Optimal energy-efficient policies for data centers through sensitivity-based optimization

Optimal energy-efficient policies for data centers through sensitivity-based optimization

On energy-aware M/G/1-LAS queue with batch arrivals

Near-optimal dispatching policy for energy-aware server clusters

Contact Info

Product

Resources

About