|
Activity Number:
|
242
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 8, 2006 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Physical and Engineering Sciences
|
| Abstract - #307613 |
|
Title:
|
A Structural Equation Method for Modeling Data Center Thermal Distribution
|
|
Author(s):
|
Zhiguang Qian*+ and Yasuo Amemiya
|
|
Companies:
|
Georgia Institute of Technology and IBM T. J. Watson Research Center
|
|
Address:
|
7 Burdick Street, Stamford, CT, 06905,
|
|
Keywords:
|
latent variable ; structural equation system ; computer experiment ; data center ; multivariate statistical analysis
|
|
Abstract:
|
Temperature management is key in designing and running a reliable data center. Physics-based mathematical models, implemented in computer code, are often used to study air movement and temperature distribution mechanisms. A statistical method based on latent variables is introduced for analyzing the multivariate temperature readings produced by the computer experiment, and for building a surrogate model to be used for prediction. A two-stage estimation procedure is developed for the proposed latent variable model by making use of sufficient statistics and ordinary least square estimation. Also discussed is a method for determining practical configurations to meet some physical and usage requirements.
|
