Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Heat: Satellite's meat is GPU's poison

Zhehu Yuan, Jinyang Liu, Guanqun Song, Ting Zhu

TL;DR

The paper investigates using GPUs as passive heaters for LEO satellites during eclipse, leveraging on-board computation to generate heat while allowing cooling during sunlit phases. It introduces a two-stage design and a fragmentation-based workload scheduling approach to precisely regulate heat output without losing computation progress. Experimental evaluation on an RTX 3070 setup contrasts execution-dominated and memory-dominated workloads, showing higher heat generation and faster temperature rise for the former, and highlighting cost and size considerations relative to traditional heaters. The study demonstrates potential advantages in cost-efficiency and adaptability, while outlining future work including real-satellite trials, improved heat transfer, and robust scheduling strategies to make GPU-based heating practical for space missions.

Abstract

In satellite applications, managing thermal conditions is a significant challenge due to the extreme fluctuations in temperature during orbital cycles. One of the solutions is to heat the satellite when it is not exposed to sunlight, which could protect the satellites from extremely low temperatures. However, heat dissipation is necessary for Graphics Processing Units (GPUs) to operate properly and efficiently. In this way, this paper investigates the use of GPU as a means of passive heating in low-earth orbit (LEO) satellites. Our approach uses GPUs to generate heat during the eclipse phase of satellite orbits, substituting traditional heating systems, while the GPUs are also cooled down during this process. The results highlight the potential advantages and limitations of this method, including the cost implications, operational restrictions, and the technical complexity involved. Also, this paper explores the thermal behavior of GPUs under different computational loads, specifically focusing on execution-dominated and FLOP-dominated workloads. Moreover, this paper discusses future directions for improving GPU-based heating solutions, including further cost analysis, system optimization, and practical testing in real satellite missions.

Heat: Satellite's meat is GPU's poison

TL;DR

The paper investigates using GPUs as passive heaters for LEO satellites during eclipse, leveraging on-board computation to generate heat while allowing cooling during sunlit phases. It introduces a two-stage design and a fragmentation-based workload scheduling approach to precisely regulate heat output without losing computation progress. Experimental evaluation on an RTX 3070 setup contrasts execution-dominated and memory-dominated workloads, showing higher heat generation and faster temperature rise for the former, and highlighting cost and size considerations relative to traditional heaters. The study demonstrates potential advantages in cost-efficiency and adaptability, while outlining future work including real-satellite trials, improved heat transfer, and robust scheduling strategies to make GPU-based heating practical for space missions.

Abstract

In satellite applications, managing thermal conditions is a significant challenge due to the extreme fluctuations in temperature during orbital cycles. One of the solutions is to heat the satellite when it is not exposed to sunlight, which could protect the satellites from extremely low temperatures. However, heat dissipation is necessary for Graphics Processing Units (GPUs) to operate properly and efficiently. In this way, this paper investigates the use of GPU as a means of passive heating in low-earth orbit (LEO) satellites. Our approach uses GPUs to generate heat during the eclipse phase of satellite orbits, substituting traditional heating systems, while the GPUs are also cooled down during this process. The results highlight the potential advantages and limitations of this method, including the cost implications, operational restrictions, and the technical complexity involved. Also, this paper explores the thermal behavior of GPUs under different computational loads, specifically focusing on execution-dominated and FLOP-dominated workloads. Moreover, this paper discusses future directions for improving GPU-based heating solutions, including further cost analysis, system optimization, and practical testing in real satellite missions.
Paper Structure (15 sections, 6 figures, 1 table, 2 algorithms)

This paper contains 15 sections, 6 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Background
  3. GPU Side
  4. Satellite Side
  5. Our Design: GPU Heater
  6. Advantages of GPU-Based Heating
  7. Challenges and Solutions
  8. Experiment Setup
  9. Execution-Dominated Workload
  10. Memory-Dominated Workload
  11. Results
  12. The Relationship between Workload and Temperature
  13. Power, Size, and Price
  14. Future Work
  15. Conclusion

Figures (6)

  • Figure 1: The two stages of GPU based on the location of the satellite
  • Figure 2: The two stages of GPU based on the location of the satellite
  • Figure 3: The two stages of GPU based on the location of the satellite
  • Figure 4: The temperature of GPU against time
  • Figure 5: The running time of each round of execution under different GPU temperature
  • ...and 1 more figures