摘要
在数字经济与人工智能深度融合的背景下,算力资源已成为科研活动的核心生产力。然而,高校算力平台建设普遍面临资源分散、利用率低等挑战。以华中科技大学为对象,提出并实践了一种服务创新驱动科研效能提升的新范式。这个范式通过“众筹式”建设模式有效整合校内分散算力资源,实现共建共享;采用容器化混合调度技术支持高性能计算(HPC)与人工智能(AI)负载的灵活部署;结合AI信息化技术打造无感化服务生态。实践表明,这个模式成功实现了资源聚合,显著降低了运维成本,提升了科研效率并产出了丰富成果,为高校算力资源的集约化管理与服务创新提供了重要的理论参考与实践指南。
Amid the deep convergence of the digital economy and artificial intelligence,computing resources have become the core productivity of scientific research activities.However,the construction of computing platforms in universities generally faces challenges such as resource fragmentation and low utilization rates.Taking Huazhong University of Science and Technology(HUST)as a case study,this paper proposes and practices a novel service innovation paradigm to enhance research efficiency.This paradigm effectively integrates fragmented on-campus computing resources through a“crowdsourced”development model,achieving co-construction and sharing.It employs containerized hybrid scheduling technology to flexibly deploy high-performance computing(HPC)and artificial intelligence(AI)workloads.Furthermore,it integrates AI-enabled information technology to create a frictionless service ecosystem.Practice demonstrates that this model successfully achieves resource aggregation,significantly reduces operational costs,enhances research efficiency,and produces abundant outcomes.It provides important theoretical references and practical guidelines for the centralized management and service innovation of computing resources in universities.
作者
张策
柳斌
ZHANG Ce;LIU Bin(Network and Informatization Office,Huazhong University of Science and Technology,Wuhan 430074)
出处
《中国科技资源导刊》
2025年第5期59-64,共6页
China Science & Technology Resources Review
关键词
科研效能
众筹式模式
算力平台建设
资源整合
容器化调度
无感化服务
research efficiency
crowdsourced model
computing platform construction
resource integration
containerized scheduling
frictionless service
