TY - JOUR

T1 - Scaling Power Management in Cloud Data Centers

T2 - A Multi-Level Continuous-Time MDP Approach

AU - Chitsaz, Behzad

AU - Khonsari, Ahmad

AU - Moradian, Masoumeh

AU - Dadlani, Aresh

AU - Talebi, Mohammad Sadegh

N1 - Publisher Copyright: IEEE

PY - 2024

Y1 - 2024

N2 - Power management in multi-server data centers especially at scale is a vital issue of increasing importance in cloud computing paradigm. Existing studies mostly consider thresholds on the number of idle servers to switch the servers on or off and suffer from scalability issues. As a natural approach in view of the Markovian assumption, we present a multi-level continuous-time Markov decision process (CTMDP) model based on state aggregation of multi-server data centers with setup times that interestingly overcomes the inherent intractability of traditional MDP approaches due to their colossal state-action space. The beauty of the presented model is that, while it keeps loyalty to the Markovian behavior, it approximates the calculation of the transition probabilities in a way that keeps the accuracy of the results at a desirable level. Moreover, near-optimal performance is attained at the expense of the increased state-space dimensionality by tuning the number of levels in the multi-level approach. The simulation results were promising and confirm that in many scenarios of interest, the proposed approach attains noticeable improvements, namely a near 50% reduction in the size of CTMDP while yielding better rewards as compared to existing fixed threshold-based policies and aggregation methods.

AB - Power management in multi-server data centers especially at scale is a vital issue of increasing importance in cloud computing paradigm. Existing studies mostly consider thresholds on the number of idle servers to switch the servers on or off and suffer from scalability issues. As a natural approach in view of the Markovian assumption, we present a multi-level continuous-time Markov decision process (CTMDP) model based on state aggregation of multi-server data centers with setup times that interestingly overcomes the inherent intractability of traditional MDP approaches due to their colossal state-action space. The beauty of the presented model is that, while it keeps loyalty to the Markovian behavior, it approximates the calculation of the transition probabilities in a way that keeps the accuracy of the results at a desirable level. Moreover, near-optimal performance is attained at the expense of the increased state-space dimensionality by tuning the number of levels in the multi-level approach. The simulation results were promising and confirm that in many scenarios of interest, the proposed approach attains noticeable improvements, namely a near 50% reduction in the size of CTMDP while yielding better rewards as compared to existing fixed threshold-based policies and aggregation methods.

KW - Cloud computing

KW - Cloud data centers

KW - Costs

KW - Data centers

KW - Delays

KW - markov decision process

KW - Power demand

KW - power management

KW - Servers

KW - setup time

KW - state aggregation

KW - Switches

UR - http://www.scopus.com/inward/record.url?scp=85182924419&partnerID=8YFLogxK

U2 - 10.1109/TSC.2024.3354202

DO - 10.1109/TSC.2024.3354202

M3 - Journal article

AN - SCOPUS:85182924419

JO - IEEE Transactions on Services Computing

JF - IEEE Transactions on Services Computing

SN - 1939-1374

ER -