In NUMA (non-uniform memory architecture) machines,
memory requests missing in LLC may be serviced either by local or remote DRAM.
Memory requests to remote DRAM incur much greater latencies than those to local
DRAM. It is recommended to keep as much frequently accessed data local as
possible. This metric is defined by the ratio of cycles when LLC load misses
are serviced by remote DRAM to all cycles.
A significant amount of time is spent servicing memory requests from
remote DRAM. Wherever possible, try to consistently use data on the same core,
or at least the same package, as it was allocated on.