From the news
- Chancellor Amma Addresses the Parliament of World’s Religions
- Amrita Students Qualify for the European Mars Rover Challenge
Publication Type : Conference Proceedings
Publisher : 2017 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS)
Source : 2017 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS), IEEE, Bhubaneswar, India (2017)
Url : https://www.scopus.com/record/display.uri?eid=2-s2.0-85049955100&origin=resultslist&sort=plf-f&src=s&sid=a8f73ff31047ba959d38d874c6453979&sot=autdocs&sdt=autdocs&sl=18&s=AU-ID%2856021058900%29&relpos=2&citeCnt=0&searchTerm=
Keywords : Cloud computing, Cloud service providers, computing resources, energy consumption, generic application, Java, Meters, openstack implementation, optimal trade-off, optimisation, performance consumption, policy-driven power management, Power Consumption, Power demand, Power Management, scale-out techniques, scale-up techniques, Servers, Virtual machining
Campus : Amritapuri
School : Centre for Cybersecurity Systems and Networks, School of Engineering
Center : Cyber Security
Department : cyber Security
Year : 2017
Abstract : Cloud computing enables users to rent computing resources on-demand towards meeting the needs of diverse applications. However, scaling of resources may incur significant impact on performance and power consumption which are the two key concerns for cloud service providers. The major goal of cloud providers is to develop policies for balancing the conflicting objectives of maximizing performance and minimizing energy consumption. Towards this goal, we analyze the impact of scale-up and scale-out techniques for varying cloud workloads through an OpenStack implementation. Our analysis reveals that these techniques vary with the nature of applications that run on the cloud as a result of which policies need to be developed on a per-application basis. We develop a threshold-based policy which determines the optimal trade-off depending on the application profile. Our proposed policy is generic and can be applied to other workloads thus facilitating efficient management of resources.
Cite this Research Publication : T. V. Ram and Sankaran, S., “Towards policy-driven power management for cloud computing”, 2017 IEEE International Conference on Advanced Networks and Telecommunications Systems (ANTS). IEEE, Bhubaneswar, India, 2017.