Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Sensitivity Analysis of Flexibility from GPU-Heavy Data Centers

Yiru Ji, Constance Crozier, Matthew Liska

Abstract

The rapid growth of GPU-heavy data centers has significantly increased electricity demand and creating challenges for grid stability. Our paper investigates the extent to which an energy-aware job scheduling algorithm can provide flexibility in GPU-heavy data centers. Compared with the traditional first-in first-out (FIFO) baseline, we show that more efficient job scheduling not only increases profit, but also brings latent power flexibility during peak price period. This flexibility is achieved by moving lower energy jobs, preferentially executing jobs with lower GPU utilization and smaller node requirements, when the electricity price is high. We demonstrate that data centers with lower queue length and higher variance in job characteristics such as job GPU utilization and job size, offer the greatest flexibility potential. Finally we show that data center flexibility is highly price sensitive, a 7% demand reduction is achieved with a small incentive, but unrealistically high prices are required to achieve a 33% reduction.

A Sensitivity Analysis of Flexibility from GPU-Heavy Data Centers

Abstract

The rapid growth of GPU-heavy data centers has significantly increased electricity demand and creating challenges for grid stability. Our paper investigates the extent to which an energy-aware job scheduling algorithm can provide flexibility in GPU-heavy data centers. Compared with the traditional first-in first-out (FIFO) baseline, we show that more efficient job scheduling not only increases profit, but also brings latent power flexibility during peak price period. This flexibility is achieved by moving lower energy jobs, preferentially executing jobs with lower GPU utilization and smaller node requirements, when the electricity price is high. We demonstrate that data centers with lower queue length and higher variance in job characteristics such as job GPU utilization and job size, offer the greatest flexibility potential. Finally we show that data center flexibility is highly price sensitive, a 7% demand reduction is achieved with a small incentive, but unrealistically high prices are required to achieve a 33% reduction.

Paper Structure

This paper contains 18 sections, 4 equations, 3 figures, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Baseline Data Center Operation
  3. Parameter definitions
  4. State-of-the art scheduling
  5. Energy-Aware Scheduling Algorithm
  6. Optimization over a static time window
  7. Rolling horizon algorithm
  8. Simulation Set Up and Flexibility Calculation
  9. Simulation set up
  10. Testing flexibility
  11. Results
  12. Overall Performance Comparison
  13. Peak Price Period Analysis
  14. Sensitive Analysis on different job characteristics
  15. GPU utilization variance
  16. ...and 3 more sections

Figures (3)

  • Figure 1: Node Occupation and GPU Utilization change over time considering peak price with a 300x multiplier.
  • Figure 2: Average power usage during peak price period time under different price multipliers.
  • Figure 3: Percentage change of average power during the peak price period compare with FIFO under different peak price and GPU variance.