Cache hierarchy and memory subsystem of the AMD opteron processor

Abstract

The 12-core AMD Opteron processor, code-named "Magny Cours," combines advances in silicon, packaging, interconnect, cache coherence protocol, and server architecture to increase the compute density of high-volume commodity 2P/4P blade servers while operating within the same power envelope as earlier-generation AMD Opteron processors. A key enabling feature, the probe filter, reduces both the bandwidth overhead of traditional broadcast-based coherence and memory latency. © 2006 IEEE.