DEVELOP0MENT OF ADVANCED COMPUTER HARDWARE AND SOFTWARE

先进计算机硬件和软件的开发

• 批准号: 3774968
• 负责人: B R BROOKS
• 金额: --
• 依托单位: CENTER FOR INFORMATION TECHNOLOGY
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/3774968
• 关键词: computer data analysis; computer program /software; computer simulation; computer system design /evaluation; computer system hardware; macromolecule; molecular dynamics; parallel processing

With the advent of new computer technology amenable to large-scale computing, software and hardware development efforts are essential for optimal use of these resources. The efforts include the developing of techniques of exploit parallel multi-machines, writing assembler code for commercial processor, and the establishing of a parallel workstation cluster for high-efficiency simulations at low cost. Development of methods and software to make productive use of parallel MIMD machines for use in macromolecular simulations is underway. The initial global communication approach has been successful in providing an efficient full-feature version of CHARMM. This parallel version of CHARMM has been extended to run on almost any MIMD parallel computer platform: Intel iPSC/860, Intel delta, CM-5, IBM/SP1, Convex SP1, and on clusters of workstations. Our current development effort involves a scalable algorithm that promises to greatly reduce the communication cost for very large MPP machines or for large workstation clusters. Current projects include: - A Scalable Molecular Dynamics Algorithm for MPP Machines and Large Workstation Clusters. - Development of Parallel QM/MM Methods - Development and Efficient Use of a High Speed Workstation Cluster of HP735s - Development and Support of CHARMM (Chemistry at HARvard Macromolecular Mechanics)

随着适合大规模计算的新计算机技术的出现 计算、软件和硬件开发工作对于 优化利用这些资源。 这些努力包括开发 利用并行多机技术，编写汇编代码 商用处理器，并行工作站的建立 集群以低成本实现高效模拟。 开发有效利用并行的方法和软件 用于大分子模拟的 MIMD 机器正在进行中。 这 最初的全球沟通方法已成功提供 CHARMM 的高效全功能版本。 这个并行版本 CHARMM 已扩展为可以在几乎所有 MIMD 并行计算机上运行 平台：Intel iPSC/860、Intel delta、CM-5、IBM/SP1、Convex SP1 和 在工作站集群上。 我们当前的开发工作涉及 可扩展的算法有望大大降低通信成本 对于非常大的 MPP 机器或大型工作站集群。 目前的项目包括： - 适用于 MPP 机器和大型机器的可扩展分子动力学算法 工作站集群。 - 并行 QM/MM 方法的开发 - 高速工作站集群的开发和高效使用 HP735s - CHARMM（哈佛高分子化学）的开发和支持 力学）