Section: New Results

Allocation algorithms in Cloud platforms

In the context of service hosting in large-scale datacenters, we provide [11] a deep analysis of a cluster data trace recently released by Google and we focus on a number of questions which have not been addressed in previous studies. In particular, we describe the characteristics of job resource usage in terms of dynamics (how it varies with time), of correlation between jobs (identify daily and/or weekly patterns), and correlation inside jobs between the different resources (dependence of memory usage on CPU usage). From this analysis, we derive scalable formalizations of the allocation problem which encompass most job features. In [19] , [22] , we study one such model, where long-running services experience demand variations with a periodic (daily) pattern. Such services account for most of the overall CPU demand. This leads to an allocation problem where the classical Bin-Packing issue is augmented with the possibility to co-locate jobs whose peaks occur at different times of the day, which is bound to be more efficient than the usual approach that consists in over-provisioning for the maximum demand. We propose mathematical formulations, column generation approaches, and analyze their performance compared to standard packing heurisics (such as Best-Fit or First-Fit Decreasing). We show that taking periodicity of demand into account allows for a substantial improvement on machine utilization in the context of large-scale, state-of-the-art production datacenters, and that column generation allows to obtain quasi-optimal solutions in reasonable time.