Biblio
Resource scheduling in a computing system addresses the problem of packing tasks with multi-dimensional resource requirements and non-functional constraints. The exhibited heterogeneity of workload and server characteristics in Cloud-scale or Internet-scale systems is adding further complexity and new challenges to the problem. Compared with,,,, existing solutions based on ad-hoc heuristics, Machine Learning (ML) has the potential to improve further the efficiency of resource management in large-scale systems. In this paper we,,,, will describe and discuss how ML could be used to understand automatically both workloads and environments, and to help to cope with scheduling-related challenges such as consolidating co-located workloads, handling resource requests, guaranteeing application's QoSs, and mitigating tailed stragglers. We will introduce a generalized ML-based solution to large-scale resource scheduling and demonstrate its effectiveness through a case study that deals with performance-centric node classification and straggler mitigation. We believe that an MLbased method will help to achieve architectural optimization and efficiency improvement.
In the wake of diversity of service requirements and increasing push for extreme efficiency, adaptability propelled by machine learning (ML) a.k.a self organizing networks (SON) is emerging as an inevitable design feature for future mobile 5G networks. The implementation of SON with ML as a foundation requires significant amounts of real labeled sample data for the networks to train on, with high correlation between the amount of sample data and the effectiveness of the SON algorithm. As generally real labeled data is scarce therefore it can become bottleneck for ML empowered SON for unleashing their true potential. In this work, we propose a method of expanding these sample data sets using Generative Adversarial Networks (GANs), which are based on two interconnected deep artificial neural networks. This method is an alternative to taking more data to expand the sample set, preferred in cases where taking more data is not simple, feasible, or efficient. We demonstrate how the method can generate large amounts of realistic synthetic data, utilizing the GAN's ability of generation and discrimination, able to be easily added to the sample set. This method is, as an example, implemented with Call Data Records (CDRs) containing the start hour of a call and the duration of the call, in minutes taken from a real mobile operator. Results show that the method can be used with a relatively small sample set and little information about the statistics of the true CDRs and still make accurate synthetic ones.