Paper Info
Title
A Two-stage Strategy to Optimize Energy Consumption for Latency-critical Workload under QoS Constraint.
Abstract
Data centers afford huge energy costs. Saving energy while providing efficient quality of service (Qos) is\r\nthe goal pursued by data centers. But this is a challenging issue. To ensure the Qos of latency-critical\r\napplications, data centers always schedule processors to run at higher frequencies. The continuous high\u0002\r\nfrequency operation will cause great energy waste. Modern processors are equipped with dynamic voltage\r\nand frequency scaling (DVFS) technology, which allows the processor to run at every frequency levels it\r\nsupports, so we focus on how to use DVFS to trade-off between energy and Qos. In this paper, we propose\r\na two-stage strategy based on DVFS to dynamically scaling the CPU frequency during latency-critical\r\nworkload execution, aimed at minimizing the energy consumption for latency-critical workload which is\r\nunder the Qos constraint. The two-stage strategy includes a static stage and dynamic stage, which are\r\nworked together to determine the optimal frequency for running workload. The static stage uses a well\u0002\r\ndesigned heuristic algorithm to determine the frequency-load matches under Qos constraint, while the\r\ndynamic stage leverages a threshold method to determine whether to adjust the pre-set frequency. We\r\nevaluate the two-stage strategy in terms of Qos and energy saving on the cloudsuite benchmark, and\r\ncompares the two metrics with the-state-of art Ondemand. Results show that our strategy is superior to\r\nOndemand for energy saving, improving more than 13%.
Year
DOI
Venue
2020
10.5755/j01.itc.49.4.25029
Inf. Technol. Control.
DocType
Volume
Issue
Journal
49
4
Citations 
PageRank 
References 
0
0.34
0
Authors
3
Name
Order
Citations
PageRank
Jingwei Li1677.10
Duanyu Teng200.34
Jinwei Lin300.68