Abstract
With the emergence of commodity computing environments (i.e. clouds), information technology (IT) infrastructure providers are creating data centers in distributed geographical regions. Since geographic regions have different costs and demands on their local power grids, cloud computing infrastructures will require innovative management procedures to ensure energy-efficiency that spans multiple regions. Macro-level measurement of energy consumption that focuses on the individual servers does not have the dynamism to respond to situations where domain-specific software services are migrated to different data centers in varying regions. Next-generation models will have to understand the impact on power consumption for a particular software application or software service, at a micro-level. A challenge to this approach is to develop a prediction of energy conservation a priori. In this work, we discuss the challenges for measuring the power consumption of an individual web service. We discuss the challenges of determining the power consumption profile of a web service each time it is migrated to a new server and the training procedure of the power model. This potentially promotes creating a dynamically-green cloud infrastructure.
| Original language | English |
|---|---|
| Title of host publication | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
| Pages | 87-96 |
| Number of pages | 10 |
| Volume | 7221 LNCS |
| DOIs | |
| State | Published - Sep 3 2012 |
| Externally published | Yes |
| Event | 2011 International Conference on Service-Oriented Computing, ICSOC 2011 - Paphos, Cyprus Duration: Dec 5 2011 → Dec 8 2011 |
Publication series
| Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
|---|---|
| Volume | 7221 LNCS |
| ISSN (Print) | 03029743 |
| ISSN (Electronic) | 16113349 |
Other
| Other | 2011 International Conference on Service-Oriented Computing, ICSOC 2011 |
|---|---|
| Country | Cyprus |
| City | Paphos |
| Period | 12/5/11 → 12/8/11 |
Fingerprint
Keywords
- Energy-awareness
- green web service
- service-oriented software engineering
- web service
ASJC Scopus subject areas
- Computer Science(all)
- Theoretical Computer Science
Cite this
Engineering energy-aware web services toward dynamically-green computing. / Bartalos, Peter; Blake, M. Brian.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7221 LNCS 2012. p. 87-96 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7221 LNCS).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Engineering energy-aware web services toward dynamically-green computing
AU - Bartalos, Peter
AU - Blake, M. Brian
PY - 2012/9/3
Y1 - 2012/9/3
N2 - With the emergence of commodity computing environments (i.e. clouds), information technology (IT) infrastructure providers are creating data centers in distributed geographical regions. Since geographic regions have different costs and demands on their local power grids, cloud computing infrastructures will require innovative management procedures to ensure energy-efficiency that spans multiple regions. Macro-level measurement of energy consumption that focuses on the individual servers does not have the dynamism to respond to situations where domain-specific software services are migrated to different data centers in varying regions. Next-generation models will have to understand the impact on power consumption for a particular software application or software service, at a micro-level. A challenge to this approach is to develop a prediction of energy conservation a priori. In this work, we discuss the challenges for measuring the power consumption of an individual web service. We discuss the challenges of determining the power consumption profile of a web service each time it is migrated to a new server and the training procedure of the power model. This potentially promotes creating a dynamically-green cloud infrastructure.
AB - With the emergence of commodity computing environments (i.e. clouds), information technology (IT) infrastructure providers are creating data centers in distributed geographical regions. Since geographic regions have different costs and demands on their local power grids, cloud computing infrastructures will require innovative management procedures to ensure energy-efficiency that spans multiple regions. Macro-level measurement of energy consumption that focuses on the individual servers does not have the dynamism to respond to situations where domain-specific software services are migrated to different data centers in varying regions. Next-generation models will have to understand the impact on power consumption for a particular software application or software service, at a micro-level. A challenge to this approach is to develop a prediction of energy conservation a priori. In this work, we discuss the challenges for measuring the power consumption of an individual web service. We discuss the challenges of determining the power consumption profile of a web service each time it is migrated to a new server and the training procedure of the power model. This potentially promotes creating a dynamically-green cloud infrastructure.
KW - Energy-awareness
KW - green web service
KW - service-oriented software engineering
KW - web service
UR - http://www.scopus.com/inward/record.url?scp=84865498058&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84865498058&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-31875-7_10
DO - 10.1007/978-3-642-31875-7_10
M3 - Conference contribution
SN - 9783642318740
VL - 7221 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 87
EP - 96
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ER -