At present, improvements of the computing nodes as well as the optimization of interconnect networks are two main approaches to boost HPC’s performance.
当前,提高大规模并行计算机的性能主要从两个方面着手:单个计算结点性能的提高以及连接各计算结点的高速互连网络的优化与改进。
参考来源 - 高性能计算机无缓存光互连网络技术研究·2,447,543篇论文数据,部分数据来源于NoteExpress
Think of a compute cluster or grid composed of thousands of commodity computing nodes.
(想一想拥有数千个计算节点的计算集群或网格)。
This example runs on two computing nodes and the targeted Teradata database has two AMPs.
本例运行于两个计算节点之上,并且目标teradata数据库有两个AMP。
All computing nodes must be built on x86 64 bit hardware supporting up to 128gb of memory, IA-64 not being supported.
所有的计算节点必须是基于64位x86的硬件,支持最高128g内存,并不支持IA - 64。
应用推荐