Abstract
There are strict requirements and needs to avoid huge downtime cost for data center cooling systems to operate continuously for reliable data center services. Current design standards achieve this by installing redundant cooling equipment. Other equipment including distribution headers is also used. However, no study is found to quantify the reliability of these designs comprehensively using on-site performance data. It is unknown how much reliability can be improved by adopting these designs or if they are burdens to the cost of systems only. This study quantifies how different data center cooling system configurations enhance the reliability and availability of data centers. The results show that it is crucial to install a redundant chiller or redundant chiller plant with alternative cooling sources to meet the requirement of high-tier data centers, but other common practices such as distribution headers and the 2(N + 1) configuration do not improve the reliability and availability of data center cooling systems effectively.
| Original language | English |
|---|---|
| Article number | 106573 |
| Journal | Reliability Engineering and System Safety |
| Volume | 191 |
| DOIs | |
| Publication status | Published - Nov 2019 |
Keywords
- Availability
- Chiller
- Cooling
- Data center
- Reliability
ASJC Scopus subject areas
- Safety, Risk, Reliability and Quality
- Industrial and Manufacturing Engineering