Thermal Management of Datacenter

1
Thermal Management of Datacenter

• Qinghui Tang

2
Preliminaries

• What is data center
• What is thermal management
• Why does Intel Care
• Why Computer Science

3
Typical layout of a datacenter

• Rack outlet temperature Tout
• Rack inlet temperature Tin
• Air conditioner supply temperature Ts

4
State-of-Art Thermal Management of Data Center

• Power densities are increasing exponentially
  along with Moores Law
• Current cooling solutions at various levels
• Chip / component level
• Server/board level
• Rack level
• Data center level
• S/W based Thermal management solutions HPDuke

5
Thermal Management of Datacenter

• Motivation and significance
• Compute Intensive Applications (Online Gaming,
  Computer Movie Animation, Data Mining) requiring
  increased utilization of Data Center
• Maximizing computing capacity is a demanding
  requirement
• New blade servers can be packed more densely
• Energy cost is rising dramatically
• Goal
• Improving thermal performance
• Lowering hardware failure rate
• Reducing energy cost

6
Typical layout of a datacenter
7
New Challenges

• Planning perspective How to design efficient
  data center?
• does upgrading 10 blade servers to smart ones
  help to reduce cost
• Operation perspective How to efficiently operate
  data center and lower the cost?
• Whats the trade-off between utility cost and
  hardware failure cost
• Overcooling wastes energy and increases utility
  cost
• Undercooling increases frequency of hardware
  failures

8
Research Issues of Thermal Management of
Datacenter
Scheduler
Other Impact Factors
Control
Thermal Performance Evaluation
Cost Optimization
Abstract Heat Flow Model
Power Load Characterization
Modeling Thermal Performance
Multiscale Multimodal Info Analysis
Understanding
9
Example of multiple granularity and scale
10
Multiscale and multimodal nature of datacenter
management

• Information perspective
• Multiple system variables
• Different change pattern
• Different sampling Rate
• Control perspective
• Responsiveness
• Control granularity (spatial and temporal level)
• Sensitivity Analysis

11
Approaches

• CFD simulation to characterize thermal
  performance of data center
• Online measurement and feedback control system

12
CFD Simulation
CFD real model based on ASU HPC center
13
Thermal-aware task scheduling
6
5
1
2
4
3
14
Two-Pronged Approach

• Real-time measurement
• Online lightweight simulation prediction

15
Goal Datacenter energy cost optimization
16
Different optimization goals

• Maximizing computation capacity given energy cost
  constraint
• Minimizing individual cost (computing
  cost/cooling cost)
• Achieving thermal balancing

17
Questions and answers