Internet Video Data Streaming by Ye Tian Min Zhao & Xinming Zhang

Author:Ye Tian, Min Zhao & Xinming Zhang
Language: eng
Format: epub
Publisher: Springer Singapore, Singapore

3.2 Energy Saving for CDN

3.2.1 Energy-Aware Load Balancing

Mathew et al. [9] also address the energy saving problem for CDN. More specifically, an energy-aware CDN should turn off content servers during periods of low workload while seeking to balance three key design goals: maximize energy reduction, minimize the impact on client-perceived service availability (SLAs), and limit the frequency of on-off server transitions to reduce wear-and-tear and its impact on hardware reliability.

The authors propose energy-aware CDN load-balancing algorithms in two stages. The local load balancing algorithm works within a CDN server cluster, and makes decisions on hibernating and waking up servers based on past and current workload, then the global energy-aware load balancing algorithm redistributes traffic across clusters according to real-time cluster capacities. Evaluation using Akamai CDN workload shows that by holding an extra 10% of the servers as live spares, the local load balancing algorithm achieves a system-wide energy reduction of 55% and a service availability of at least five nine’s (99.999%), while incurring an average of at most 1 transition per server per day. Although having limited impact on energy reduction, the global load balancing algorithm can reduce 10–25% server transitions by redistributing workload across proximal clusters.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.