Computational Core

计算建模核心将专注于构建、支持和扩展计算 Cobre Pi‘s用于生物材料合成和表征的模式。核心总监，在 与核心顾问一起监督核心的日常运作，长期规划 可持续发展和增长、用户培训和支持、软件/硬件管理和教师 在使用现有资源的情况下进行指导/指导。Cobre集群将与 NSF MRL支持的GPU(图形处理单元)集群位于化学和生物化学中 部门。在此Cobre的过程中，我们将计划继续增长和更新当前 通过替换为1)具有更快互连的高密度(大核数)节点和2) 更快、自适应、可扩展的存储容量。我们预计新教职员工的用户使用量会增加， 而核心将通过员工咨询为用户提供培训，核心总监与用户互动， 并在校园内开设了许多课程来利用这些设施。 计算建模核心的拟议具体目标是： 1)提供可靠、灵活的异类计算基础设施(硬件、软件、存储、 联网)支持生物材料设计、合成和表征；2)扩展计算能力 利用图形处理单元(GPU)进行生物材料建模的能力；3)开发强大的 存储和归档基础架构，可适应不断增加的数据存储负载 硬件和其他用户；以及4)为教师、研究生、博士后提供培训 当代和新兴高性能计算领域的研究人员和本科生 范例。

The Computational Modeling Core will focus on building, supporting, and extending the computational modalities used by COBRE Pi's for biomaterials synthesis and characterization. The Core Director, in conjunction with the Core Advisor will oversee daily operations of the Core, long-term planning for sustainability and growth, user training and support, and software/hardware management, and faculty mentoring/guidance in the context of using available resources. The COBRE cluster will integrate with the NSF MRl supported GPU (Graphics Processing Unit) cluster housed in the Chemistry and Biochemistry Department. During the course of this COBRE, we will plan to continue growing and updating the current cluster through replacement with 1) high-density (large core count) nodes with faster interconnects and 2) faster, adaptive, extensible storage capacities. We will anticipate increased user usage with newer faculty, and the CORE will support users with training via staff consultation, CORE director interactions with users, and numerous courses on campus to utilize the facilities. Proposed specific aims for the Computational Modeling Core are: 1) to provide reliable, flexible heterogeneous computational infrastructure (hardware, software, storage, networking) to support biomaterials design, synthesis, and characterization; 2) to extend Computational capabilities to exploit Graphics Processing Units (GPU's) for biomaterials modeling; 3) to develop a robust storage and archival infrastructure accommodating increasing data-storage loads incurred with faster hardware and additional users; and 4) to supply training for faculty, graduate students, post-doctoral researchers, and undergraduate students in contemporary and emerging high-performance computing paradigms.