Finding the appropriate CPU frequency by using PESOS Algorithm
Author(s):
B.Ashok Kumar, S.Vani
Keywords:
Web search engine,Predictive Energy Saving Online Scheduling Algorithm (PESOS), CPU, Energy.
Abstract
A web search engine may be a code that's designed to go looking for data on the World Wide Web. The search results are typically bestowed in a very line of results usually spoken as search engine results pages (SERPs). Internet search engines are composed by thousands of question process nodes, i.e., servers dedicated to method user queries. Such several servers consume a big quantity of energy, principally responsible to their CPUs, however they're necessary to make sure low latencies, since users expect sub-second response times. However, users will hardly notice response times that ar quicker than their expectations. Hence, we tend to propose the predictive Energy Saving online scheduling algorithm (PESOS) to pick the foremost acceptable frequency to process a question on a per-core basis. PESOS aims at method queries by their deadlines, and leverage high-level programming data to reduce the energy consumption of a question processing node. PESOS bases its call on question potency predictors, estimating the process volume and interval of a question. We through an experiment appraise PESOS upon the TREC Clue Web09B assortment and also the MSN2006 question log. Results show that PESOS will cut back the energy consumption of a question processing node up to ∼48% compared to a system running at most C.P.U. core frequency.
Article Details
Unique Paper ID: 145519
Publication Volume & Issue: Volume 4, Issue 10
Page(s): 504 - 507
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024