DiRAC: Memory Intensive 2.5x

天文学、核物理学和粒子物理学界的物理学家们都专注于了解宇宙是如何在一个非常基本的层面上运作的。他们工作的距离尺度变化了50个数量级，从大型强子对撞机实验探测到的最小距离，原子核深处，到太空中发现的最大尺度的星系团。然而，科学挑战是通过诸如宇宙是如何开始的以及它是如何演变的问题联系在一起的。宇宙的基本组成和结构是什么，它们是如何相互作用的？进步需要新的天文观测和实验数据，但也需要新的理论见解。理论理解越来越多地来自大规模计算，这使我们能够非常准确地用数据来面对我们理论的后果，或者让我们能够详细地询问数据以提取对我们理论有影响的信息。这些计算测试了我们拥有的最快的计算机，并推动了这一领域的技术边界。DiRAC 2 HPC设施自2012年以来一直在运行，为STFC支持的粒子物理、天文学、宇宙学和核物理等所有领域的理论研究提供计算资源。它是一个高生产力的设施，每年在国际同行评审期刊上产生250多篇论文。然而，DiRAC 2硬件现在至少有5年的历史，因此有很大的失败风险。任何一项DiRAC 2服务的丢失都将对依赖它来提供科学研究的研究社区产生潜在的灾难性影响。DiRAC 2.5x项目申请资金的主要目的是更换达勒姆老化的DiRAC 2硬件，爱丁堡和莱斯特同时利用最近的硬件进步来提供一些新的功能（例如使用闪存的i/o加速）作为建议的DiRAC 3服务的原型。DiRAC-2.5x项目建立在DiRAC-2.5 HPC设施，并将提供所需的资源，以支持尖端研究从1/4/2018开始，在所有科学领域的支持STFC。特别是资金排序由达勒姆将允许：达勒姆可运行的计算量增加2倍，可用计算能力增加50%（假设当前的DiRAC-2.5系统继续以当前水平运行）。该系统的使用将主要由DiRAC资源分配委员会决定，但预计增强的系统将用于非常大的计算，例如：（i）模拟产生引力波的黑洞对的合并，例如LIGO财团最近发现的引力波; ㈡对宇宙中星系的形成和演变进行迄今为止最真实的模拟对太阳内部和行星内部进行详细的模拟。

Physicists across the astronomy, nuclear and particle physics communities are focussedon understanding how the Universe works at a very fundamental level. The distance scaleswith which they work vary by 50 orders of magnitude from the smallest distances probedby experiments at the Large Hadron Collider, deep within the atomicnucleus, to the largest scale galaxy clusters discovered out in space. The Science challenges,however, are linked through questions such as: How did the Universe begin and how is it evolving?and What are the fundamental constituents and fabric of the Universe and how do they interact?Progress requires new astronomical observations and experimental data but alsonew theoretical insights. Theoretical understanding comes increasingly from large-scalecomputations that allow us to confront the consequences of our theories very accuratelywith the data or allow us to interrogate the data in detail to extract information that hasimpact on our theories. These computations test the fastest computers that we have andpush the boundaries of technology in this sector. They also provide an excellentenvironment for training students in state-of-the-art techniques for code optimisation anddata mining and visualisation.The DiRAC2 HPC facility has been operating since 2012, providingcomputing resources for theoretical research in all areas of particlephysics, astronomy, cosmology and nuclear physics supported bySTFC. It is a highly productive facility, generating more than 250papers annually in international, peer-reviewed journals. However, theDiRAC2 hardware is now at least 5 years old and is therefore atsignificant risk of failure. The loss of any one of the DiRAC2services would have a potentially disastrous impact on the researchcommunities which rely on it to deliver their scientific research. Themain purpose of the requested funding for the DiRAC2.5x project is toreplace the ageing DiRAC2 hardware at Durham, Edinburgh and Leicesterwhile taking advantage of recent hardware advances to provide some newcapabilities (e.g. i/o acceleration using flash storage) as prototypesfor the proposed DiRAC3 services.The DiRAC-2.5x project builds on the success of the DiRAC-2.5 HPC facility and will provide the resources neededto support cutting edge research starting from 1/4/2018 in all areas of science supported by STFC.Specifically the funding sort by Durham will allow: A factor 2 increase in the size of calculation that can be run at Durham, and a 50% increase in theavailable computing power (assuming the current DiRAC-2.5 systems continue to operate at the current level). The usage of the system will be decided by the DiRAC Resource Allocation Committee primarily,but it is envisaged that the enhanced system will be used for very large calculations, for example, to:(i) simulate the merger of pairs of black holes which generate gravitational waves such as those recently discovered by theLIGO consortium;(ii) perform the most realistic simulations to date of the formation and evolution of galaxies in the Universe(iii) carry out detailed simulations of the interior of the sun and of planetary interiors.