Saving the planet with bin packing - Experiences using 2D and 3D bin packing of virtual machines for greener clouds
Journal article, Peer reviewed
Permanent lenke
https://hdl.handle.net/10642/3205Utgivelsesdato
2014Metadata
Vis full innførselSamlinger
Originalversjon
Hage, T., Begnum, K., & Yazidi, A. (2014, December). Saving the planet with bin packing-Experiences using 2D and 3D bin packing of virtual machines for greener clouds. In Cloud Computing Technology and Science (CloudCom), 2014 IEEE 6th International Conference on (pp. 240-245). IEEE. http://dx.doi.org/10.1109/CloudCom.2014.155Sammendrag
Greener cloud computing has recently become anextremely pertinent research topic in academy and amongpractitioners. Despite the abundance of the state of the artstudies that tackle the problem, the vast majority of themsolely rely on simulation, and do not report real settingsexperience. Thus, the theoretical models might overlook someof the practical details that might emerge in real life scenarios.In this paper, we try to bridge the aforementioned gap inthe literature by devising and also deploying algorithms forsaving power in real-life cloud environments based on variantsof the 2D/3D bin packing algorithms. The algorithms aretested on a large OpenStack deployment in use by staff andstudents at Oslo and Akershus University College, Norway. Wepresent three different adaptions of 2D and 3D bin packing,incorporating different aspects of the cloud as constraints.Our real-life experimental results show that although thethree algorithms yield a decrease in power consumption, theydistinctly affect the way the cloud has to be managed. A simplebin packing algorithm provides useful mechanism to reducepower consumption while more sophisticated algorithms do notmerely achieve power savings but also minimize the numberof migrations.