When processors access some data, a copy is made in their local cache, but space remains allocated in the home node.
Compared to NUMA, this reduces the number of redundant copies and may allow more efficient use of the memory resources.
Hardware memory coherence mechanisms are typically used to implement the migration.
Hybrid NUMA-COMA organizations have also been proposed, such as Reactive NUMA, which allows pages to start in NUMA mode and switch to COMA mode if appropriate and is implemented in the Sun Microsystems's WildFire.
[1][2] A software-based Hybrid NUMA-COMA implementation was proposed and implemented by ScaleMP,[3] allowing for the creation of a shared-memory multiprocessor system out of a cluster of commodity nodes.