• Energy Research

Cloud infrastructures are designed to simultaneously service many, diverse applications that consist of collections of Virtual Machines (VMs). The placement policy used to map applications onto physical servers has important effects in terms of application performance and resource efficiency. We propose enhancing placement policies with network-aware optimizations, trying to simultaneously improve application performance, resource efficiency and power efficiency. The per-application placement decision is formulated as a bi-objective optimization problem (minimizing communication cost and the number of physical servers on which an application runs) whose solution is searched using evolutionary techniques. We have tested three multi-objective optimization algorithms with problem-specific crossover and mutation operators. Simulation-based experiments demonstrate how, in comparison with classic placement techniques, a low-cost optimization results in improved assignments of resources, making applications run faster and reducing the energy consumed by the data center. This is beneficial for both cloud clients and cloud providers.

Load forecasting is crucial for multiple energy management tasks such as scheduling generation capacity, planning supply and demand, and minimizing energy trade costs. Such relevance has increased even more in recent years due to the integration of renewable energies, electric cars, and microgrids. Conventional load forecasting techniques obtain single-value load forecasts by exploiting consumption patterns of past load demand. However, such techniques cannot assess intrinsic uncertainties in load demand, and cannot capture dynamic changes in consumption patterns. To address these problems, this paper presents a method for probabilistic load forecasting based on the adaptive online learning of hidden Markov models. We propose learning and forecasting techniques with theoretical guarantees, and experimentally assess their performance in multiple scenarios. In particular, we develop adaptive online learning techniques that update model parameters recursively, and sequential prediction techniques that obtain probabilistic forecasts using the most recent parameters. The performance of the method is evaluated using multiple datasets corresponding with regions that have different sizes and display assorted time-varying consumption patterns. The results show that the proposed method can significantly improve the performance of existing techniques for a wide range of scenarios. \c{opyright} 2021 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works