Toward Strategic Demand Management of Public Supercomputing Resources: Policy Gaps and Institutional Improvements
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This study aims to propose strategic improvements to Korea’s supercomputer demand management policy. Through a comprehensive review of the existing legal framework and the operational plans of national and specialized centers, key structural limitations were identified and categorized into three areas: demand forecasting and planning, resource construction, and ecosystem development. To address these challenges, the study proposes a set of policy measures, including the establishment of mid- to long-term supply-demand planning mechanisms, the incorporation of resource quality metrics, resource provisioning based on reserve ratios, and the formalization of supercomputing within the national science and technology classification system. These recommendations collectively provide a foundation for transitioning from a supply-centric to a demand-responsive governance model in high-performance computing policy.
Abstract
This study aims to propose strategic improvements to Korea’s supercomputer demand management policy. Through a comprehensive review of the existing legal framework and the operational plans of national and specialized centers, key structural limitations were identified and categorized into three areas: demand forecasting and planning, resource construction, and ecosystem development. To address these challenges, the study proposes a set of policy measures, including the establishment of mid- to long-term supply-demand planning mechanisms, the incorporation of resource quality metrics, resource provisioning based on reserve ratios, and the formalization of supercomputing within the national science and technology classification system. These recommendations collectively provide a foundation for transitioning from a supply-centric to a demand-responsive governance model in high-performance computing policy.
참고문헌 (13)
- Al Faisal, Faiz, "HFBN: An energy efficient high performance hierarchical interconnection network for exascale supercomputer", IEEE Access, 10, pp.3088~3104 (2021)
- Bartolini, Andrea, "Unveiling Eurora—Thermal and power characterization of the most energy-efficient supercomputer in the world", 2014 Design, Automation & Test in Europe Conference & Exhibition (DATE) (2014), IEEE
- Hamzaoui, Ikhlasse, "A survey on the current challenges of energy-efficient cloud resources management", SN Computer Science, 1, pp.1~28 (2020)
- Jeyaraj, Pandia Rajan, "Computer‐assisted demand‐side energy management in residential smart grid employing novel pooling deep learning algorithm", International Journal of Energy Research, 45(5), pp.7961~7973 (2021)
- Madhoo, H., "Future internet concepts for demand management", 2015 International Conference on the Domestic Use of Energy (DUE) (2015), IEEE
- 3rd Master Plans for Fostering of National Supercomputing, "Ministry of Science and ICT", ~ (2023)
- Mir, Aneeque A., "A review of electricity demand forecasting in low and middle-income countries: The demand determinants and horizons", Sustainability, 12(15), pp.5931~ (2020)
- Okazaki, Ryohei, "Supercomputer Fugaku Cpu A64fx realizing high performance, high-density packaging, and low power consumption", Fujitsu Technical Review (2020)
- Quichiz, Luis Palacios, "It demand management in organizations: a review", Proceedings of the 8th International Conference on Computer Modeling and Simulation (2017)
- Savin, G. I., "Joint supercomputer center of the Russian Academy of Sciences: Present and future", Lobachevskii Journal of Mathematics, 40, pp.1853~1862 (2019)
- Shim, Hyungwook, "A study on demand management plans for National Supercomputer resources", Technology in Society, 75, pp.102376~ (2023)
- Tomčala, Jiří, "Supercomputer power consumption prediction using machine learning, nonlinear algorithms, and statistical methods", Journal of Physics: Conference Series, 2090(1) (2021), IOP Publishing