The study of datacenter location is an active research area. Popular Internet services are hosted in geographically distributed datacenters and their locations has a direct impact on the Internet services response time and costs (fixed and variable). The selection of Datacenter location takes into account several criteria including proximity to market areas, energy sources in the region, landed costs, electricity, workforce and the average temperature in the region (related to the cooling of equipment). Our objectives are to minimise datacenter costs and to calculate the power usage effectiveness metric. Our project introduces an efficient model based on mixed integer linear programming (MILP) for the datacenter localization problem. The proposed model is used to locate datacenters under Internet services demands and, at the same time, ensure a low energy consumption of information technology (IT) equipment. The resolution of the MILP model is done with an optimization solver: LINGO. Model validation and results are proved via experimentation example and numerical study.
| Date | 4 Jun 2015 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Mustapha Ouhimmou (Supervisor) & Marc Paquet (Co-supervisor) |
|---|
Ounifi, H. A. (Author),
Ouhimmou (Supervisor) &
Paquet (Co-supervisor),
4 Jun 2015Student thesis: Master's thesis › Master in Engineering: Engineering