skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Efficiency and Reliability Analyses of AC and 380V DC Distribution in Data Centers

Journal Article · · IEEE Access
 [1];  [2];  [2];  [3];  [4];  [2]
  1. Amazon AWS, Seattle, WA (United States)
  2. South Dakota State Univ., Brookings, SD (United States)
  3. Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
  4. Microsoft Inc., Redmond, WA (United States)

Modern data centers consume large amounts of electricity, resulting in high operational costs. The efficiency of a data center power distribution system can be increased, and the operational cost reduced, if the number of power conversion stages can be minimized and more efficient converters, such as wide bandgap (WBG) converters, are used. This can be achieved by using DC distribution voltage at the rack level by eliminating extra conversion stages. In this paper, benchmarks for both AC and 380V DC data centers were developed and efficiency analyses were performed for an entire year. The impact of integrating photovoltaic (PV) systems into the data centers has also been analyzed in both cases in terms of efficiency. The results show that 380V DC data centers are more efficient than AC data centers with and without PV integration. Furthermore, the reliability of an AC system was compared to a 380V DC architecture with Tier-IV standard. Monte-Carlo simulations were used to perform reliability analyses for different levels of redundancy in the Uninterruptible Power Supply (UPS) system for both cases. In conclusion, the simulation results showed that the 380V DC distribution system had a higher level of reliability than the AC distribution system in data centers.

Research Organization:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Organization:
USDOE
Grant/Contract Number:
AC05-76RL01830
OSTI ID:
1482212
Report Number(s):
PNNL-SA-138940
Journal Information:
IEEE Access, Vol. 6; ISSN 2169-3536
Publisher:
IEEECopyright Statement
Country of Publication:
United States
Language:
English
Citation Metrics:
Cited by: 21 works
Citation information provided by
Web of Science