Model Identification for Energy-Aware Management of Web Service Systems

Authors:
Mara Tanelli;Danilo Ardagna;Marco Lovera;Li Zhang
Affiliations:
Dipartimento di Elettronica e Informazione, Politecnico di Milano, Milano, Italy 20133 and Dipartimento di Ingegneria dell'Informazione e Metodi Matematici, Università degli studi di Bergamo, ...;Dipartimento di Elettronica e Informazione, Politecnico di Milano, Milano, Italy 20133;Dipartimento di Elettronica e Informazione, Politecnico di Milano, Milano, Italy 20133;IBM Research, T.J. Watson Research Center, Yorktown Heights NY 10598
Venue:
ICSOC '08 Proceedings of the 6th International Conference on Service-Oriented Computing
Year:
2008

Citing 8
Cited 1

Identification of the deterministic part of MIMO state space models given in innovations form from input-output data

Automatica (Journal of IFAC) - Special issue on statistical signal processing and control
Managing energy and server resources in hosting centers

SOSP '01 Proceedings of the eighteenth ACM symposium on Operating systems principles
Performance Guarantees for Web Server End-Systems: A Control-Theoretical Approach

IEEE Transactions on Parallel and Distributed Systems
Theory, Volume 1, Queueing Systems

Theory, Volume 1, Queueing Systems
Load prediction models in web-based systems

valuetools '06 Proceedings of the 1st international conference on Performance evaluation methodolgies and tools
SLA based resource allocation policies in autonomic environments

Journal of Parallel and Distributed Computing
Analytic modeling of multitier Internet applications

ACM Transactions on the Web (TWEB)
Bound analysis of closed queueing networks with workload burstiness

SIGMETRICS '08 Proceedings of the 2008 ACM SIGMETRICS international conference on Measurement and modeling of computer systems

An adaptive model-free resource and power management approach for multi-tier cloud environments

Journal of Systems and Software

Quantified Score

Hi-index	0.01

Visualization

Abstract

In SOA environments, service providers need to comply with the service level objectives stipulated in contracts with their customers while minimizing the operating costs of the physical infrastructure, mainly related to energy costs. The problem can be effectively formalized by using system identification and control theory: the service levels are translated into set-points for the response times of the hosted applications, and performance are traded-off with energy saving objectives based on suitable models for server dynamics. As the behavior of the incoming workload changes significantly within a single business day, control-oriented system identification approaches are very promising to model such systems, especially at a very fine grained time scales and in transient conditions. In this paper Linear Parameter Varying (LPV) state space system identification algorithms are analyzed for modeling Web services systems. The suitability of LPV models is investigated and their performance assessed by experimental data.