#AMD #EPYC #Bergamo is Fantastically Fresh Take on #CloudNative Compute
AMD was able to shrink each #CCD by adopting new #Zen4c core. AMD has less L3 cache (half at 2MB versus 4MB) with Zen 4c, and that allowed company to shrink area in a straightforward manner. Big change is transitioning to 12x 8-core CCDs on 96-core #Genoa to 8x 16-core CCD on Bergamo. Expect #EPYC9754 to be 15-20% faster than #EPYC9654 not a straight 33% as would be expected due to core count increase
https://www.servethehome.com/amd-epyc-bergamo-epyc-9754-cloud-native-sp5/