Optimal energy-efficient policies for data centers through sensitivity-based optimization
From MaRDI portal
Publication:2177776
DOI10.1007/S10626-019-00293-XzbMATH Open1435.68056arXiv1808.07905OpenAlexW2982452500WikidataQ126868249 ScholiaQ126868249MaRDI QIDQ2177776FDOQ2177776
Authors: Jing-Yu Ma, Li Xia, Quan-Lin Li 
Publication date: 6 May 2020
Published in: Discrete Event Dynamic Systems (Search for Journal in Brave)
Abstract: 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.
Full work available at URL: https://arxiv.org/abs/1808.07905
Recommendations
- Algorithms for energy conservation in heterogeneous data centers
- Energy-efficient strategy with a speed switch and a multiple-sleep mode in cloud data centers
- Energy-aware data centers with \(s\)-staggered setup and abandonment
- Energy-Efficient Server Clusters
- Impacts of Retrials on Power-Saving Policy in Data Centers
Queues and service in operations research (90B22) Performance evaluation, queueing, and scheduling in the context of computer systems (68M20) Markov and semi-Markov decision processes (90C40)
Cites Work
- Title not available (Why is that?)
- Performance optimization of queueing systems with perturbation realization
- Generalized inverses and their application to applied probability problems
- Stochastic learning and optimization. A sensitivity-based approach.
- \(M/G/k\) with staggered setup
- M/M/1 Queueing Decision Processes with Monotone Hysteretic Optimal Policies
- Exact analysis of the \(\mathrm{M}/\mathrm{M}/k/\mathrm{setup}\) class of Markov chains via recursive renewal reward
- Constructive computation in stochastic models with applications. The RG-factorization.
- Two Types ofRG-Factorizations of Quasi-birth-and-death Processes and Their Applications to Stochastic Integral Functionals
- Event-based optimization of admission control in open queueing networks
- Structural results for the control of queueing systems using event-based dynamic programming
- A Jackson network model and threshold policy for joint optimization of energy and delay in multi-hop wireless networks
- Max-Min Optimality of Service Rate Control in Closed Queueing Networks
- A tutorial on event-based optimization -- a new optimization framework
- Managing performance and power consumption in a server farm
- Service rate control of closed Jackson networks from game theoretic perspective
- Parameterized Markov decision process and its application to service rate control
- Optimal Operation of an M/M/2 Queue with Removable Servers
- Optimal service‐capacity allocation in a loss system
- Optimal Control of State-Dependent Service Rates in a MAP/M/1 Queue
- Dynamic Pricing Control for Open Queueing Networks
- Multiserver queues with finite capacity and setup time
- Assigning multiple job types to parallel specialized servers
- Group-server queues
- Instruction-throughput regulation in computer processors with data-center applications
- Energy-performance trade-off for processor sharing queues with setup delay
- Service Rate Control of Tandem Queues With Power Constraints
- Decision Processes with Monotone Hysteretic Policies
Cited In (9)
- Impacts of Retrials on Power-Saving Policy in Data Centers
- Algorithms for energy conservation in heterogeneous data centers
- Algorithms for energy conservation in heterogeneous data centers
- Energy-efficient operation of indirect adiabatic data center cooling systems via Newton-like phasor extremum seeking control
- An overview for Markov decision processes in queues and networks
- A complete algebraic solution to the optimal dynamic rationing policy in the stock-rationing queue with two demand classes
- The optimal dynamic rationing policy in the stock-rationing queue
- Optimal dynamic mining policy of blockchain selfish mining through sensitivity-based optimization
- Sensitivity-based optimization for blockchain selfish mining
This page was built for publication: Optimal energy-efficient policies for data centers through sensitivity-based optimization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2177776)
