PROJECT TITLE :

Robust Workload and Energy Management for Sustainable Data Centers

ABSTRACT:

A giant variety of geo-distributed knowledge centers begin to surge in the time of knowledge deluge and info explosion. To meet the growing demand in huge data processing, the infrastructure of future data centers should be energy-economical and sustainable. Facing this challenge, a systematic framework is put forth during this paper to integrate renewable energy sources (RES), Distributed Storage units, cooling facilities, in addition to dynamic pricing into the workload and energy management tasks of a information center network. To address RES uncertainty, the resource allocation task is formulated as a robust optimization drawback minimizing the worst-case net cost. Compared with existing stochastic optimization ways, the proposed approach entails a deterministic uncertainty set where generated RES reside, therefore can be readily obtained in follow. It is further shown that the problem will be cast as a convex program, and then solved in an exceedingly distributed fashion using the twin decomposition method. By exploiting the spatio-temporal diversity of native temperature, workload demand, energy costs, and renewable availability, the proposed approach outperforms existing alternatives, as corroborated by in depth numerical tests performed using real data.