Abstract
This paper addresses a process-to-machine reassignment problem arising in cloud computing environments. The problem formulation has been posed as the ROADEF/EURO challenge 2012. Our presented approach is basically a large neighborhood search that iteratively improves a given solution. In each iteration only a subset of processes is considered for reassignment and the new assignments are evaluated by a constraint program. In this paper we present our general solution approach. Furthermore, we evaluate different process selection strategies and other optimization means to improve the performance on larger instances. In addition, we present a simple way to compute tight lower bounds of the necessary costs.
Similar content being viewed by others
Notes
Technically the current search node only holds a reference to the base values of the root search node and a small object containing the changes that happened since then.
In practice the cost for placing \(p\) to \(m\) is cached and only updated if other processes have been placed on \(m\) in between.
References
Ahuja, R., Ergun, O., Orlin, J., & Punnen, A. (2002). A survey of very large-scale neighborhood search techniques. Discrete Applied Mathematics, 123(13), 75–102. doi:10.1016/S0166-218X(01)00338-9.
Armbrust, M., Fox, A., Griffith, R., Joseph, A. D., Katz, R., Konwinski, A., et al. (2010). A view of cloud computing. Communications of the ACM, 53(4), 50–58.
Artigues, C., Bourreau, E., Asfar, M., Kedad-Sidhoum, S., & Gueré, E. (2012). Google ROADEF/EURO challenge 2012: Machine reassignment. http://challenge.roadef.org/2012/.
Beck, J., & Siewiorek, D. (1996). Modeling multicomputer task allocation as a vector packing problem. In 9th international symposium on system synthesis, 1996. Proceedings.
Bessiere, C., Hebrard, E., Hnich, B., Kiziltan, Z., & Walsh, T. (2006). Filtering algorithms for the nvalue constraint. Constraints, 11(4), 271–293. doi:10.1007/s10601-006-9001-9.
Brandt, F., Speck, J., & Völker, M. Constraint-based very large-scale neighborhood search for machine reassignment. https://github.com/fbrandt/ROADEF2012-J25.
Cao, J., Wu, Y., & Li, M. (2012). Energy efficient allocation of virtual machines in cloud computing environments based on demand forecast. In R. Li, J. Cao, & J. Bourgeois (Eds.), Advances in grid and pervasive computing, Lecture notes in computer science (Vol. 7296, pp. 137–151). Berlin: Springer.
Masson, R., Vidal, T., Michallet, J., Penna, P. H. V., Petrucci, V., Subramanian, A., et al. (2012). An iterated local search heuristic for multi-capacity bin packing and machine reassignment problems. Technical report CIRRELT-2012-70. Interuniversity Research Centre on Enterprise Networks, Logistics and Transportation, Montreal, Canada.
Mehta, D., O’Sullivan, B., & Simonis, H. (2012). Comparing solution methods for the machine reassignment problem. Technical report. Cork Constraint Computation Centre, University College Cork, Cork, Ireland. http://4c.ucc.ie/~hsimonis/reassignment.
Moreno, I. S., Yang, R., Xu, J., & Wo, T. (2013) Improved energy-efficiency in cloud datacenters with interference-aware virtual machine placement. In 2013 IEEE eleventh international symposium on autonomous decentralized systems (ISADS) (pp. 1–8).
Panigrahy, R., Talwar, K., Uyeda, L., & Wieder, U. (2011). Heuristics for vector bin packing. Technical report. Microsoft Research.
Petrucci, V., Carrera, E., Loques, O., Leite, J. C. B., & Mosse, D. (2011). Optimized management of power and performance for virtualized heterogeneous server clusters. In 2011 11th IEEE/ACM international symposium on cluster, cloud and grid computing (CCGrid) (pp. 23–32).
Piao, J. T., & Yan, J. (2010). A network-aware virtual machine placement and migration approach in cloud computing. In 2010 9th international conference on grid and cooperative computing (GCC) (pp. 87–92).
Schulte, C., Tack, G., & Lagerkvist, M. Z. (2013). Modeling and programming with Gecode. www.gecode.org/doc-latest/MPG.
Verma, A., Ahuja, P., & Neogi, A. (2008). pmapper: Power and migration cost aware application placement in virtualized systems. In V. Issarny & R. Schantz (Eds.), Middleware 2008, Lecture notes in computer science (Vol. 5346, pp. 243–264). Berlin: Springer.
Wilcox, D., McNabb, A., Seppi, K., & Flanagan, K. (2010). Probabilistic virtual machine assignment. In The first international conference on cloud computing, GRIDs, and virtualization (pp. 54–60).
Author information
Authors and Affiliations
Corresponding author
Appendix: Additional tables
Appendix: Additional tables
In the experimental section, we used solution scores instead of the underlying solution values to compare different approaches and techniques. This decision was motivated by the fact that solution scores enable the reader to easily relate our solution to the best solution that was computed by any contestant. Additionally, a score-based evaluation is less instance-dependent and thus makes it easier to compare the performance of different strategies with each other. Nevertheless, we also want to present the actual costs of the computed solutions in order to enable others to compare their solution costs directly with the solutions computed by our approach. For this reason, Table 9 states the average costs of the solutions that were computed by our approach after 1, 5 and 30 min as well as the initial cost and our computed lower bound.
Rights and permissions
About this article
Cite this article
Brandt, F., Speck, J. & Völker, M. Constraint-based large neighborhood search for machine reassignment. Ann Oper Res 242, 63–91 (2016). https://doi.org/10.1007/s10479-014-1772-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10479-014-1772-6