Traditionally, a workstation's throughput depends on its bus and memory architecture, as well as its CPU speed.
传统上讲，工作站的吞吐量与其总线和存储器体系结构以及CPU的速度有关。
NUMA alleviates these bottlenecks by limiting the number of CPUs on any one memory bus and connecting the various nodes by means of a high speed interconnection.
NUMA通过限制任何一条内存总线上的CPU数量并依靠高速互连来连接各个节点，从而缓解了这些瓶颈状况。
The dual-ported RAM is a kind of special memory, and the bus data share between double high-speed microprocessors is carried out by using it.
双端口ram是一种特殊的数据存储芯片，利用双端口ram可以实现双高速单片机总线方式的数据共享。
应用推荐