MRI Collaborative: Development of a Data-Intensive Scalable Computing Instrument for High Performance Computing

方案编号：13-38078PI(s): Chen, Yong；格罗普，威廉·d；史密斯，菲利普W.；太阳,Xian-He;院校：德州理工大学职称：MRI协作/开发项目建议：该项目开发DICSI，一种全面的计算仪器，弥补了现有以计算为中心的高性能计算仪器在数据密集型应用方面的局限性，支持高性能计算系统设计、计算化学、生物技术和大气科学等五个大型研究项目。该项目在引入应用感知和解耦执行范式概念的基础上，解决了研究原型与工程解决方案之间的巨大差距。预计该仪器将对未来的应用、算法和仪器设计产生影响，因为它可以在支持数据密集型科学方面开辟新的研究领域，并可能重塑国家计算设施和一些机构采用的HPC仪器。除了传统的HPC计算节点外，DISCI还有一组专门设计的数据节点。数据节点提供原位数据处理，以减少数据移动和数据访问延迟，并在必要时作为“胖”计算节点进行动态配置，而计算节点的功能与传统仪器保持相同。这些数据节点与计算节点协同工作，它们一起为数据密集型HPC提供最佳的系统性能。基于软硬件协同开发原则，该仪器由DISCI系统架构和DISCI运行时软件两部分组成。系统架构构建了一个以数据为中心的HPC仪器。运行时软件扩展了MPI（消息传递接口）和MPI- io库，以支持数据节点及其相关的原位处理。该仪器将支持和促进化学动力学模拟、湍流模拟、大气数据同化和天气预报、计算生物学以及pi和高级人员进行的计算机系统等领域的研究活动。更广泛的影响：该开发项目将使学术部门、跨学科单位、组织和多组织协作能够整合他们的开发、教育和推广工作。为吸引代表性不足的学生参与DISCI的发展，院校会与有关组织的院校项目协调。所获得的经验将整合到本科和研究生课程以及夏季培训中，以使学生参与到发展中来。该教育计划的重点是支持数据密集型高性能计算，并培训一支具有广泛包容性和全球竞争力的科学劳动力队伍。该项目预计将为未来的国家高性能计算仪器提供途径，以支持数据密集型科学。此外，它可能对构建百亿亿次高性能计算仪器产生直接影响。

Proposal #: 13-38078PI(s): Chen, Yong; Gropp, William D.; Smith, Philip W.; Sun, Xian-He; Zhuang, YuInstitution: Texas Tech University Title: MRI Collab/Dev.: Data Intensive Scalable Computing Instrument for High Performance ComputingProject Proposed:This project, developing DICSI, an all-around computing instrument that compensates the limitations of existing computing-centric HPC instruments toward data-intensive applications, supports five large research projects in HPC system design, computational chemistry, biotechnology, and atmospheric science. Based on research introducing the application-aware and decoupled-execution paradigm concept, the project addresses the big gap between research prototypes and the engineering solution. Impact on future applications, algorithms, and instruments design is expected since the instrument could open up new research areas in supporting data-intensive sciences and possibly reshape HPC instruments adopted in National Computing Facilities and some institutions. In addition to the conventional HPC compute nodes, DISCI has a set of specially designed data nodes. The data nodes offer in-situ data processing to reduce data movement and data-access delay and dynamic provisioning as 'fat' compute nodes when necessary, while the functionality of compute nodes remains the same as in conventional instrumentation. These data nodes work with compute nodes in concert and together they provide an optimum system performance for data-intensive HPC. Based on a hardware-software co-development principle, the instrument consists of two components: the DISCI system architecture and the DISCI runtime software. The system architecture builds an HPC instrument with a data-centric view. The runtime software extends the MPI (Message Passing Interface) and MPI-IO library to support data nodes and their associated in-situ processing. The instrument will enable and foster research activities in the areas of chemical dynamics simulation, simulations of turbulent flows, atmospheric data assimilation and weather forecasting, computational biology, and computer systems that PIs and senior personnel conduct. Broader Impacts: This development project will enable academic departments, cross-disciplinary units, organizations, and multi-organization collaborations to integrate their development, education, and outreach efforts. To attract underrepresented students into the DISCI development, the institution will coordinate with institutional projects at respective organizations. The experience gained will be integrated into undergraduate and graduate courses and summer orientation trainings to get students involved in the development. The education plan concentrates on supporting data-intensive HPC and training a broadly inclusive and globally competitive science workforce. The project is expected to provide the pathway to future national HPC instruments to support data-intensive sciences. Furthermore, it could have a direct impact on building exascale HPC instruments.