Chameleon adapting throughput server to time-varying green power budget using online learning

Eco-friendly energy sources (i.e. green power) attract great attention as lowering computer carbon footprint has become a necessity. Existing proposals on managing green energy powered systems show sub-optimal results since they either use rigid load power capping or heavily rely on backup power. We...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings of the 2013 International Symposium on Low Power Electronics and Design s. 100 - 105
Hlavní autoři: Li, Chao, Wang, Rui, Goswami, Nilanjan, Li, Xian, Li, Tao, Qian, Depei
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: Piscataway, NJ, USA IEEE Press 04.09.2013
Edice:ACM Conferences
Témata:
Applied computing > Physical sciences and engineering > Electronics
Computer systems organization > Architectures > Distributed architectures > Client-server architectures
Computer systems organization > Dependable and fault-tolerant systems and networks
Computing methodologies > Machine learning
General and reference > Cross-computing tools and techniques > Performance
Information systems > Data management systems > Middleware for databases > Application servers
Information systems > Data management systems > Middleware for databases > Database web servers
Networks > Network performance evaluation
throughput server
adaptation
learning
green power
ISBN:1479912352, 9781479912353
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Eco-friendly energy sources (i.e. green power) attract great attention as lowering computer carbon footprint has become a necessity. Existing proposals on managing green energy powered systems show sub-optimal results since they either use rigid load power capping or heavily rely on backup power. We propose Chameleon, a novel adaptive green throughput server. Chameleon comprises of multiple flexible power management policies and leverages learning algorithm to select the optimal operating mode during runtime. The proposed design outperforms the state-of-the-art approach by 13% on performance, improves system MTBF by 42%, and still maintains up to 95% green energy utilization.
ISBN:1479912352
9781479912353
DOI:10.5555/2648668.2648693