Abstract-Current data centers usually operate under poor utilization due to resource fragmentation. The hierarchical nature of data centers places a limit on the achievable aggregate bandwidth in the backbone. Suboptimal virtual machine placement also introduces unnecessary cross network traffic. In this paper, we solve a joint tenant (i.e., server or virtual machine) placement and route selection problem by exploiting multipath routing capability and dynamic virtual machine migration. These two complementary degrees of freedom: placement and routing, are mutually-dependent, and their joint optimization turns out to substantially improve data center efficiency. We propose (i) an offline algorithm that solves a static problem given a network snapshot, and (ii) an online solution for a dynamic environment with changing traffic. Leveraging and expanding the technique of Markov approximation, we propose an efficient online algorithm that requires a very small number of virtual machine migrations. Performance evaluation that employs the synthesized data center traffic traces, on various topologies and under a spectrum of elephant and mice workloads, demonstrates a consistent and significant improvement over the benchmark achieved by common heuristics used in today's data centers.

No abstract

We develop DeepOPF as a Deep Neural Network (DNN) approach for solving security-constrained direct current optimal power flow (SC-DCOPF) problems, which are critical for reliable and cost-effective power system operation. DeepOPF is inspired by the observation that solving the SC-DCOPF problem for a given power network is equivalent to depicting a highdimensional mapping between load inputs and generation and phase-angle outputs. We first construct and train a DNN to learn the mapping between the load inputs and the generations. We then directly compute the phase angles from the generations and loads by using the (linearized) power flow equations. Such a two-step procedure significantly reduces the dimension of the mapping to learn, subsequently cutting down the size of the DNN and the amount of training data/time needed. We further characterize a condition that allows us to tune the size of our neural network according to the desired approximation accuracy of the load-to-generation mapping. Simulation results of IEEE test cases show that DeepOPF always generates feasible solutions with negligible optimality loss, while speeding up the computing time by up to 400x as compared to a state-of-the-art solver.1 There are two types of SC-DCOPF problems, namely the preventive SC-DCOPF problem and the corrective SC-DCOPF problem. In the preventive SC-DCOPF problem, the system operating decisions cannot change once they are determined, thus they need to guarantee feasibility under both the preand post-contingency constraints. For the corrective SC-DCOPF problem, the system operator can have a short time (e.g., 5 minutes) [12] to adjust the operating points after the occurrence of each contingency. Our DeepOPF approach is applicable to both problems. We focus on the preventive SC-DCOPF problem in this paper for easy illustration.

Microgrids represent an emerging paradigm of future electric power systems that can utilize both distributed and centralized generations. Two recent trends in microgrids are the integration of local renewable energy sources (such as wind farms) and the use of co-generation (i.e., to supply both electricity and heat). However, these trends also bring unprecedented challenges to the design of intelligent control strategies for microgrids. Traditional generation scheduling paradigms rely on perfect prediction of future electricity supply and demand. They are no longer applicable to microgrids with unpredictable renewable energy supply and with co-generation (that needs to consider both electricity and heat demand). In this paper, we study online algorithms for the microgrid generation scheduling problem with intermittent renewable energy sources and co-generation, with the goal of maximizing the cost-savings with local generation. Based on the insights from the structure of the offline optimal solution, we propose a class of competitive online algorithms, called CHASE (Competitive Heuristic Algorithm for Scheduling Energy-generation), that track the offline optimal in an online fashion. Under typical settings, we show that CHASE achieves the best competitive ratio among all deterministic online algorithms, and the ratio is no larger than a small constant 3. We also extend our algorithms to intelligently leverage on limited prediction of the future, such as near-term demand or wind forecast. By extensive empirical evaluations using real-world traces, we show that our proposed algorithms can achieve near offline-optimal performance. In a representative scenario, CHASE leads to around * The first two authors are in alphabetical order.20% cost reduction with no future look-ahead, and the cost reduction increases with the future look-ahead window.

No abstract