1.The main thing done in the relational database was to use "soft NUMA" and port mapping to get a good distribution of work within the system.
我们在关系型数据库中完成了被称为“SoftNUMA”的技术，它通过端口映射在系统内部得以获得良好的分布式工作效果。
2.All traffic enters through a single port and is distributed on a round-robin basis to any available NUMA node.
所有通信流量都通过一个单独的端口输入并分布到任何可用的NUMA节点。
3.To understand how pages of memory from the buffer cache are assigned when using NUMA, see Growing and Shrinking the Buffer Pool Under NUMA.
若要了解使用NUMA时如何分配缓冲区高速缓存中的内存页，请参阅使用NUMA扩展和收缩缓冲池。
4.Systems with a large number of processors may find it advantageous to recompile against the NUMA user-land API's added in RHEL4.
在拥有大量处理器的系统中，可能会发现借助RHEL4中所增加的NUMA用户空间API进行重新编译会有好处。
5.NUMA, like SMP, allows users to harness the combined power of multiple processors, with each processor accessing a common memory pool.
numa与smp相似，让用户能驾驭多个处理器结合起来的能力，每个处理器能存取一个公共的存储器组。
6.NUMA reduces the contention for a system's shared memory bus by having more memory buses and fewer processors on each bus.
NUMA通过在每个总线使用更多内存总线和更少处理器来减少系统共享内存总线的争用。
7.Any operation running on a single NUMA node can only use buffer pages from that node .
针对单个NUMA节点执行的任何操作都只能使用该节点中的缓冲区页。
8.The ratio of the cost to access foreign memory over that for local memory is called the NUMA ratio.
访问外部内存的开销与访问本地内存的开销比率称为NUMA比率。
9.For high-end machines, new features target performance improvements, scalability, throughput, and NUMA support for SMP machines.
对高端的机器来说，新特性针对的是性能改进、可扩展性、吞吐率，以及对SMP机器NUMA的支持。
10.The number of CPUs within a NUMA node depends on the hardware vendor.
NUMA节点中的CPU数量取决于硬件供应商。