Intel introduced its Xeon 6 processor family featuring new P-core and E-core SKUs for workloads spanning from compute-intensive high-performance computing (HPC) and AI to scalable cloud-native applications.
The new Intel Xeon 6 processors bring generation-on-generation improvements for double data rate (DDR5), peripheral component interconnect express 5.0 (PCIe 5.0), ultra path interconnect (UPI), and compute express link (CXL):
The 6700 series offers up to 1.4x higher memory bandwidth—with multiplexer combined ranks (MCR) dual in-line memory module (DIMM) memory in P-core—to handle larger amounts of data at once and up to 1.1x increase in input/output (I/O) bandwidth versus 5th Gen Intel Xeon processors, delivering a faster, more efficient delivery system for data input and output.
The 6900 series offers up to 1.8x increased inter-socket bandwidth versus 5th Gen Xeon processors. This allows faster, more efficient communication between different parts of the system, significantly boosting performance, especially for the most demanding tasks.
Both the 6700 and 6900 series support CXL 2.0 – Type 1, Type 2, and Type 3. This new standard makes it easier for computers to connect and communicate with additional components like accelerators, memory expanders, and other devices.
Xeon 6 processors with E-cores are optimized for network and edge workloads and enable users to achieve up to 2.7x higher 5G user plane function performance per watt and up to 3.5x higher next-generation firewall performance per watt—as tested with Intel Ethernet 800 Series—compared to 2nd Gen Xeon processors. This helps to increase overall compute efficiency while supporting confidential computing capabilities for 5G networks and edge-to-cloud infrastructures.
Source: https://convergedigest.blogspot.com/2024/06/intel-unveils-xeon-6-processors-for-hpc.html