ccNUMA

ccNUMA stands for Cache Coherent Non-Uniform Memory Access. It is a type of computer memory architecture used in multiprocessor systems. In ccNUMA, each processor has its own local memory, and also access to memory that is local to other processors. The key characteristic is that the time it takes for a processor to access memory varies depending on whether the memory is local to the processor or remote (belonging to another processor). This is where the "non-uniform" part of the name comes from.

The "cache coherent" aspect refers to the mechanism that ensures all processors have a consistent view of memory, even when data is cached locally by multiple processors. When a processor modifies data in its local cache, the ccNUMA system ensures that other processors either see the updated data or are made aware that their cached copy is stale. This coherence is crucial for correct program execution in shared-memory multiprocessor systems.

ccNUMA systems are designed to scale to a larger number of processors than traditional Uniform Memory Access (UMA) systems. In UMA, all memory is equally accessible to all processors, which can create a bottleneck as the number of processors increases. ccNUMA distributes memory, reducing contention on a single shared memory bus. This architecture is commonly found in servers and high-performance computing systems. While offering scalability, ccNUMA introduces complexity in managing memory access patterns and ensuring optimal performance, as programmers and operating systems need to be aware of the memory locality to minimize latency.