NUMA-Aware Process Placement

You will optimize for memory latency by binding threads to specific nodes in a non-uniform memory access architecture.