Core C: Computation and Software Development Core

核心C：计算和软件开发核心 核心C的总体目标是为参与计划项目的研究人员提供 支持和指导他们在开发代码时的所有计算需求，以及 以科学软件的形式协助算法的实施。因此，Core C 将在每个研究项目的成功和影响中发挥核心作用。我们有 将核心活动组织为两个主要目标：高性能计算[HPC]和 科学代码开发。在这样做的过程中，我们意识到两个关键概念：可靠性和 可扩展性。 我们的第一个具体目标是高性能计算。核心将参与 活动范围从开发和分配用于计算的资源到积极审查 或者开发新的方法来处理可能既“大”又复杂的数据。具体来说， 我们将重点介绍并行计算、云计算技术和内存不足的使用 数据分析技术。 我们的第二个具体目标是科学代码开发。软件工具是一个角落- 当涉及到传播和广泛使用新的统计方法时，石头成分。 核心将协助成员开发这类软件的代码，并特别强调 关于算法的有效实现，基于两个基本原则，可靠性和 可扩展性。为此，我们将制定科学软件的标准和指导方针 开发--包括文档、版本控制、持续集成等主题 和单元测试--由项目成员遵循。此外，核心C将同时建议和 积极参与每个项目制定的方法的实施过程。 最终，核心将参与整个过程，从初始算法一直到 设计和实施的方法，直到最终的软件设计、打包和公开 放手。 此外，核心将扮演许多其他关键角色，以支持其他计划项目 组件。这些例子包括资源分配、测试数据的模拟以及 开发计算方法教程，可与 核心D的教育努力。

Core C: Computation and Software Development Core The overall goal of Core C is to provide researchers involved with the Program Project with support and guidance for all their computational needs while developing their code, as well as assistance with implementation of algorithms in the form of scientific software. As such, Core C will play a central role in the success and impact of each of the research projects. We have organized the Core activities into two main aims: High-Performance Computing [HPC] and Scientific Code Development. In doing so, we are conscious of two key concepts: Reliability, and Scalability. Our first Specific Aim focuses on high-performance computing. The Core will be involved in activities ranging from developing and allocating resources for computation, to actively reviewing or developing new methods for dealing with data that may be both ‘big’ and complex. Specifically, we will focus on the use of parallel computing, cloud computing technologies, and out-of-memory data analysis techniques. Our second Specific Aim focuses on scientific code development. Software tools are a corner- stone component when it comes to the diffusion and wide-spread use of novel statistical methods. The Core will assist members in developing the code for such software, with a special emphasis on the efficient implementation of algorithms, based on two fundamental principles, reliability and scalability. To this end, we will develop standards and guidelines on scientific software development--including themes such as documentation, version control, continuous integration and unit tests--to be followed by the project members. Furthermore, Core C will both advise and actively participate in the implementation process of the methods developed by each project. Ultimately, the Core will be involved throughout the process all the way from initial algorithmic design and implementation of the methods, up to final software design, packaging and public release. Additionally, the Core will play a number of other key roles supporting the other Program Project components. Examples of these include resource allocation, simulation of test data, and development of computational-methods tutorials that can be distributed in partnership with the educational efforts of Core D.