Non-Uniform Memory Access Model (NUMA)

In shared memory multiprocessor systems, local memories are able to be connected with every processor. The collection of all local memories forms the universal memory being shared. In this manner global memory is distributed to all processors. In this case the access to a local memory is consistent for its related processor as it is connected to the local memory. But if one reference is to local memory of another remote processor, then the access isn't uniform. It depends on the location of memory. So, all memory words are not accessed consistently.