Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Virtual machine consolidation using constraint-based multi-objective optimization
Terra-Neves M., Lynce I., Manquinho V. Journal of Heuristics25 (3):339-375,2019.Type:Article
Date Reviewed: Aug 8 2019

As cloud computing grows, its demands on resources play an ever-increasing role in the economics of machine maintenance. One potential way of addressing this issue is by improving usage. For example, by judiciously placing the tasks requested of it, a system could effectively reduce the number of physical machines required to support the demands made of the system. In some cases this could be facilitated by moving tasks in order to free up machines.

This paper considers the problem of optimizing load demands on a cloud computing system. The primary goal being to reduce the system’s energy demands. To model this, the authors describe a multi-objective Boolean encoding for this virtual machine (VM) consolidation problem. After setting up the problem, the authors describe a guided improvement algorithm (GIA) for multi-objective pseudo Boolean optimization (MOPBO). While this algorithm will produce a number of Pareto optimal placements, it is not feasible for practical use due to the time that it requires. The authors therefore describe an approximation for the algorithm, as well as a model enumeration algorithm (called ENUM in the paper) that lists solutions to the VM consolidation with migrations problem (VMCwM).

The algorithms were run on subsets of workload traces randomly selected from the Google cluster-data project (https://github.com/google/cluster-data). It should be noted that a budgetary constraint can be used to restrict the amount of task moving that takes place--this prevents excessive task moving. The benchmarks included instances with 32, 64, and 128 servers, which were compared with state-of-the-art algorithms based on stochastic methods. ENUM outperformed the remaining algorithms, with GIA coming in second place. One consideration that is only tangentially addressed is the cost of these algorithms in terms of their demands on the system.

Reviewer:  J. P. E. Hodgson Review #: CR146646 (1911-0398)
Bookmark and Share
  Featured Reviewer  
 
General (I.0 )
 
 
Heuristic Methods (I.2.8 ... )
 
Would you recommend this review?
yes
no
Other reviews under "General": Date
A multi-modal approach for determining speaker location and focus
Siracusa M., Morency L., Wilson K., Fisher J., Darrell T.  Multimodal interfaces (Proceedings of the 5th international conference, Vancouver, British Columbia, Canada, Nov 5-7, 2003)77-80, 2003. Type: Proceedings
Mar 1 2004
Nanotechnology: science and computation (Natural Computing Series)
Chen J., Jonoska N., Rozenberg G., Springer-Verlag New York, Inc., Secaucus, NJ, 2006.  393, Type: Book (9783540302957)
Aug 2 2007
High performance computing for big data: methodologies and applications
Wang C., CRC Press, Inc., Boca Raton, FL, 2018.  286, Type: Book (978-1-498783-99-6), Reviews: (1 of 2)
Apr 4 2019
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy