Politecnico di Torino (logo)

Load management in green data centers

Damiano Scantamburlo

Load management in green data centers.

Rel. Michela Meo, Daniela Renga. Politecnico di Torino, Corso di laurea magistrale in Ict For Smart Societies (Ict Per La Società Del Futuro), 2020

PDF (Tesi_di_laurea) - Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives.

Download (12MB) | Preview

Data centres usage and impact is a sensible matter, even more nowadays. Data centres energy impact is not negligible, driven by continuous technologies improvement and the research of higher performances. Among several methods to improve efficiency, this thesis work main goal is to study energy and cost impact of a proper and more tailored load management approach. The thesis aims to implement and extend a previous work born by a collaboration between University of Calabria and Polytechnics of Turin. The case study remains the same and it is represented by 4 hypothetical data centres placed in different locations. This spatial dislocation implies different energy cost and different renewable production among the 4 structure at the same hour of the day. This work, and this thesis as well, studies the efficiency gain brought by performing a load migration between these 4 data centres in order to match better renewables production and hourly energy cost. The load moved consisted in a set of virtual machines referring to real data from a Telecom data centre collected by researchers of University of Calabria. For this purpose, a java program was implemented to properly simulate the migration of virtual machines over this 4 data centre giving back energy parameters results. This thesis continues to analyse these performances making further considerations and implementations focusing on the influence of a proper virtual machine characterization. The virtual machine features considered are principally cpu, ram and disk usage. The analysis use these metrics as constraints to migration, this permits to calculate eventual benefits based on the main characteristic of each virtual machine which results evaluated individually. The output is an overall consideration on performances, sum of the 4 data centres results. The final goal is to understand and contextualize these results to find relations between the virtual machine characteristics and an optimal migration policy.

Relators: Michela Meo, Daniela Renga
Academic year: 2020/21
Publication type: Electronic
Number of Pages: 84
Corso di laurea: Corso di laurea magistrale in Ict For Smart Societies (Ict Per La Società Del Futuro)
Classe di laurea: New organization > Master science > LM-27 - TELECOMMUNICATIONS ENGINEERING
Aziende collaboratrici: UNSPECIFIED
URI: http://webthesis.biblio.polito.it/id/eprint/16613
Modify record (reserved for operators) Modify record (reserved for operators)