Thermal, management for high performance computing in spaceborne applications

被引:1
|
作者
Samson, JR [1 ]
Cutting, FM [1 ]
机构
[1] Honeywell Space Syst, Clearwater, FL 33764 USA
关键词
D O I
10.1109/ITHERM.2000.866832
中图分类号
O414.1 [热力学];
学科分类号
摘要
Power is a threefold problem for mobile electronic applications - generation, dissipation, and junction temperature. Of the three, dissipation is the most limiting for spaceborne applications. Advances in high performance computing and high density packaging exacerbates the power dissipation problem and, hence, the junction temperature problem. In many space applications, particularly on smaller space vehicles, articulated space radiator systems with sealed coolant loops, pumps, etc, are neither practical nor available. Hence, the ultimate limitation is the ability to radiate heat to other sinks - primarily and ultimately to deep space. This paper considers the limitation of deep space as a heat sink for given vehicle parameters such as maximum operating temperatures, available radiating surface area, and emissivity. Different techniques of vehicle cooling are reviewed with a brief discussion of the pros and cons of each technique. Consideration of current and future material availability with better thermal and electrical conductivity is presented. Better thermal conductivity tends to isothermalize the radiating area, thereby increasing radiant heat transfer proportional to the fourth power of the absolute temperature. Increased electrical conductivity reduces power consumption and hence the need for heat dissipation. Such considerations are included in projections of self-dissipation capability per circuit board using advanced composite materials. The direct radiation approach is addressed, including a brief discussion of the conflict between the need for spacing for thermal considerations versus the desire to keep the length of signal paths as short as possible and elimination of the beneficial effects of the spacecraft and chassis shielding. A comparison of power and thermal constraints for a 10 GFLOPS onboard processing system is traded off against current and future material availability, power dissipation, radiation surface area available, and the projected performance density of candidate onboard processing architectures using COTS components.
引用
收藏
页码:247 / 254
页数:8
相关论文
共 50 条
  • [31] THE APPLICATIONS AND TRENDS OF HIGH PERFORMANCE COMPUTING IN FINANCE
    Li Hong
    Lu Zhong-hua
    Chi Xue-bin
    PROCEEDINGS OF THE NINTH INTERNATIONAL SYMPOSIUM ON DISTRIBUTED COMPUTING AND APPLICATIONS TO BUSINESS, ENGINEERING AND SCIENCE (DCABES 2010), 2010, : 193 - 197
  • [32] Allocation management solutions for high performance computing
    Jackson, SM
    PDPTA '05: Proceedings of the 2005 International Conference on Parallel and Distributed Processing Techniques and Applications, Vols 1-3, 2005, : 10 - 16
  • [33] License management in grid and high performance computing
    Raekow, Yona
    Simmendinger, Christian
    Kraemer-Fuhrmann, Ottmar
    COMPUTER SCIENCE-RESEARCH AND DEVELOPMENT, 2009, 23 (3-4): : 275 - 281
  • [34] A review on soft computing and nanofluid applications for battery thermal management
    Can, Aykut
    Selimefendigil, Fatih
    Oztop, Hakan F.
    JOURNAL OF ENERGY STORAGE, 2022, 53
  • [35] A review on soft computing and nanofluid applications for battery thermal management
    Can, Aykut
    Selimefendigil, Fatih
    Oztop, Hakan Fehmi
    JOURNAL OF ENERGY STORAGE, 2024, 95
  • [36] Building a high performance computing clusters to use in computing course applications
    Aydin, Semra
    Bay, Omer Faruk
    WORLD CONFERENCE ON EDUCATIONAL SCIENCES - NEW TRENDS AND ISSUES IN EDUCATIONAL SCIENCES, 2009, 1 (01): : 2396 - 2401
  • [37] High-Throughput Computing Versus High-Performance Computing for Groundwater Applications
    Fienen, Michael N.
    Hunt, Randall J.
    GROUNDWATER, 2015, 53 (02) : 180 - 184
  • [38] Performance of commercial analog multiplexers for spaceborne applications
    Lum, GK
    Kinell, DK
    May, RJ
    Robinette, LE
    IEEE TRANSACTIONS ON NUCLEAR SCIENCE, 1996, 43 (06) : 3068 - 3075
  • [39] Thermal and power challenges in high performance computing systems
    Natarajan, Venkat
    Deshpande, Anand
    Solanki, Sudarshan
    Chandrasekhar, Arun
    Japanese Journal of Applied Physics, 2009, 48 (5 PART 3) : 011 - 05
  • [40] Thermal and Power Challenges in High Performance Computing Systems
    Natarajan, Venkat
    Deshpande, Anand
    Solanki, Sudarshan
    Chandrasekhar, Arun
    JAPANESE JOURNAL OF APPLIED PHYSICS, 2009, 48 (05)