Tijsma, Y.W. (2019) Tomte: modelling environmental conditions of a data centre by means of a hardware scale model. Master's Thesis / Essay, Computing Science.
|
Text
thesis_tijsma.pdf Download (21MB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (99kB) |
Abstract
Data centres have become the foundation of digital infrastructure, because they serve a wide range of activities across government, business, and society. Optimization of energy efficiency is a major focus of data centre operators in order to reduce the operational costs and to adhere to governmental guidelines concerning greenhouse gas emissions. Environmental control is often used as one lever to adjust energy consumption. Most researchers are limited to software simulators to assess the performance of novel technologies and practices to further improve the energy efficiency. However, software simulators do not model the environmental conditions. Alternatives are actual data centres and scale models, like Parasol with a cost of 300.000 dollars, but they are prohibitively expensive to use. This work proposes a novel design of a hardware scale model confined to rack level of a data centre costing a fraction of exisiting hardware scale models. This scale model coined Tomte model is able to replay data centre workloads by means of CPU utilization and output quantitative measures for temperature, humidity, and power consumption. Based on a data set from the RUG’s high performance cluster Peregrine, we replayed a 6-day workload on Tomte. Pearson correlation analysis shows that Tomte captures analogous relations between workload, environmental conditions, and power usage which are present in the Peregrine data set. Moreover, we ran the same workload using the software simulator Cl
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Supervisor name: | Lazovik, A. and Blaauw, F.J. |
| Degree programme: | Computing Science |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 29 Jan 2020 12:28 |
| Last Modified: | 29 Jan 2020 12:28 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/21463 |
Actions (login required)
 |
View Item |