MRI: Acquistion of A Scalable Instrument for Discovery through Computing

提案号：CNS 08-21527PI(s): Raghavan, Padma Chen, Long-Quing；彼得·j·哈德森；甘地，马哈茂德·t；机构：宾夕法尼亚州立大学大学公园，PA 16802-700标题：MRI/Acq。: Acq。通过计算获得可扩展的发现仪器这项来自主要研究仪器计划（MRI）的奖励为宾夕法尼亚州立大学获得一台万亿级先进计算仪器提供了资金。该仪器将使来自七个学科（生物、材料和社会科学、计算机和信息科学、工程、教育和地球科学）的研究人员能够通过计算进行发现和设计的虚拟实验。研究项目涉及：传染病动力学的预测网络建模，设计新的压电材料，设计下一代芯片多处理器，模拟人类互动以促进虚拟社区的学习，以及开发关键区域环境观测站。尽管这些项目各不相同，但它们都面临着计算可扩展性的挑战，需要解决潜在的科学进步，这些进步往往依赖于解决具有足够细节和复杂性的大型问题。该工具将构成多学科协作环境的核心，以实现变革性方法，以应对多层次扩展的挑战。它将支持一套综合研究、教育、培训和推广活动：(i)通过将扩展方法从一个领域转移到另一个领域，实现跨项目的协作扩展，同时解决单个项目中的算法、系统或仪器扩展挑战，（ii）通过工业伙伴关系促进技术转移，以及（iii）增长和增强有限的计算科学人才库的多样性。

Proposal #: CNS 08-21527PI(s): Raghavan, Padma Chen, Long-Quing; Hudson, Peter J.; Kandemir, Mahmut T.; Smith, Brian K.Institution: Pennsylvania State University University Park, PA 16802-700Title: MRI/Acq.: Acq.of A Scalable Instrument for Discovery through ComputingMRI Acquisition of a Scalable Instrument for Discovery through ComputingThis award from the Major Research Instrumentation Program (MRI)provides funds for the acquisition of a terascale advanced computinginstrument at the Pennsylvania State University. The instrumentwill enable researchers from seven disciplines (biological, materialsand social sciences, computer and information science, engineering,education, and geosciences), to perform virtual experiments towarddiscovery and design through computing. Research projects concern:predictive network modeling of infectious disease dynamics, designingnew piezoelectric materials, designing next-generation chipmultiprocessors, modeling human interactions to promote learningin virtual communities, and the development of a critical zoneenvironmental observatory. Despite their diversity, these projectsshare computational scalability challenges to be addressed forenabling scientific advances that often depend on solving largeproblems representing a sufficient level of detail and complexity.The instrument will form the core of a multidisciplinary collaborativeenvironment to enable transformative approaches to address thechallenges of scaling at multiple levels. It will support a set ofintegrated research, education, training, and outreach activitiesto: (i) enable collaborative scaling across projects through thetransfer of scaling approaches from one domain into another, whileaddressing algorithmic, system, or instrument scaling challengeswithin individual projects, (ii) promote technology-transfer throughindustrial partnerships, and (iii) grow and enhance the diversityof the limited computational science talent pool.