论文部分内容阅读
2 0 1 2年L H C实验发现希格斯粒子后,我国粒子物理学界规划了高能环形正负电子对撞机(CEPC),以及它升级后的超级质子对撞机(Spp C),用于研究希格斯物理,并在高能量前沿探索超越标准模型的新物理。CEPC实验的计算量非常大,即使是在预研阶段,一次探测器全模拟就需要833万CPU小时的计算量和1PB存储空间。当本地计算资源不足时,分布式计算是一种快速整合合作单位计算资源来满足实验大量计算需求的有效方法。2012年,高能所基于DIRAC中间件建立了一套约2000 CPU核规模的分布式计算系统,服务于BESIII实验。本文介绍了将此分布式计算系统扩展到支持CEPC实验的改进方案,并介绍了此系统在CEPC实验的实际应用。
After the LHC experiment was discovered in 2012, the particle physics community in China planned a high-energy annular positron-electron collider (CEPC) and its upgraded Super Proton Collider (Spp C) for research Higgs Physics explores new physics beyond the standard model at a high-energy frontier. CEPC experiments are computationally intensive and require 8,330,000 CPU hours of computation and 1 PB of storage for a full probe, even during the pre-research phase. When local computing resources are scarce, distributed computing is an effective way to quickly integrate the computing resources of the cooperating units to meet the massive computing needs of the experiment. In 2012, High Energy set up a distributed computing system of about 2,000 CPU cores based on DIRAC middleware to serve BESIII experiments. This article describes an extension of this distributed computing system to support CEPC experiments and describes the practical application of this system in CEPC experiments.