In this paper, we present a constraint programming approach for the service consolidation problem that is being currently tackled by Neptuny, Milan. The problem is defined as: Given a data-center, a set of servers with a priori fixed costs, a set of services or applications with hourly resource utilizations, find an allocation of applications to servers while minimizing the data-center costs and satisfying constraints on the resource utilizations for each hour of the day profile and on rule-based constraints defined between services and servers and amongst different services. The service consolidation problem can be modelled as an Integer Linear Programming problem with 0-1 variables, however it is extremely difficult to handle large sized instances and the rule-based constraints. So a constraint programming approach using the Comet programming language is developed to assess the impact of the rule-based constraints in reducing the problem search space and to improve the solution quality and scalability. Computational results for realistic consolidation scenarios are presented, showing that the proposed approach is indeed promising. © 2010 Springer-Verlag.
A Constraint Programming Approach for the Service Consolidation Problem
Gualandi, Stefano
;
2010-01-01
Abstract
In this paper, we present a constraint programming approach for the service consolidation problem that is being currently tackled by Neptuny, Milan. The problem is defined as: Given a data-center, a set of servers with a priori fixed costs, a set of services or applications with hourly resource utilizations, find an allocation of applications to servers while minimizing the data-center costs and satisfying constraints on the resource utilizations for each hour of the day profile and on rule-based constraints defined between services and servers and amongst different services. The service consolidation problem can be modelled as an Integer Linear Programming problem with 0-1 variables, however it is extremely difficult to handle large sized instances and the rule-based constraints. So a constraint programming approach using the Comet programming language is developed to assess the impact of the rule-based constraints in reducing the problem search space and to improve the solution quality and scalability. Computational results for realistic consolidation scenarios are presented, showing that the proposed approach is indeed promising. © 2010 Springer-Verlag.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
