4th Gen Intel® Xeon® Scalable processors |
Introducing 4th Gen Intel® Xeon® Scalable processors, designed to accelerate performance across the fastest-growing workloads. These processors have the most built-in accelerators of any CPU on the market to help improve performance efficiency for emerging workloads, especially those powered by AI. In addition to performance improvements, 4th Gen Intel® Xeon® Scalable processors have advanced security technologies to help protect data in an ever-changing landscape of threats while unlocking new opportunities for business insights.
Artificial Intelligence: Up to | Networking & Storage: Up to | Data Analytics: Up to | ||
10x | 2x | 3x | ||
higher PyTorch performance for both real-time inference and training with built-in Intel AMX (BF16) versus the prior generation (FP32).1 | the capacity at the same power envelope for vRAN workloads versus prior-generation processors.2 | higher RocksDB performance using integrated Intel IAA versus the prior generation.3 |
PCI Express Gen 5 |
| DDR5 | CXL | |
Unlock new I/O speeds with opportunities to enable the highest possible throughput between the CPU and connected devices. | Improve compute performance by overcoming data bottlenecks with higher memory bandwidth. | Reduce compute latency in the data center and help lower TCO with CXL 1.1 for next-generation workloads. | ||
| ||||
Up to 8-socket scalability |
| Up to 4-socket scalability |
| Up to 2-socket scalability |
Four Intel UPI ports at 16 GT/s |
| Three Intel UPI ports at 16 GT/s |
| Two Intel UPI ports at 16 GT/s |
80 lanes of PCIe 5.0 with CXL |
| 80 lanes of PCIe 5.0 with CXL |
| 80 lanes of PCIe 5.0 with CXL |
DDR5 at up to 4,800 MT/s (1 DIMM per channel) or 4,400 MT/s (2 DIMMs per channel) |
| DDR5 at up to 4,800 MT/s (1 DIMM per channel) or 4,400 MT/s (2 DIMMs per channel) |
| DDR5 at up to 4,800 MT/s (1 DIMM per channel) or 4,400 MT/s (2 DIMMs per channel) |
Intel® Optane™ persistent memory (PMem) 300 series supported |
| Intel Optane PMem 300 series supported |
| - |
Intel AVX-512 (two 512-bit FMAs) |
| Intel AVX-512 (two 512-bit FMAs) |
| Intel AVX-512 (two 512-bit FMAs) |
Intel® Hyper-Threading Technology (Intel® HT Technology) and Intel® Turbo Boost Technology |
| Intel HT Technology and Intel Turbo Boost Technology |
| Intel HT Technology and Intel Turbo Boost Technology |
Intel AMX |
| Intel® Deep Learning Boost (Intel® DL Boost) and Intel AMX |
| Intel DL Boost and Intel AMX |
Intel SST |
| Intel SST |
| - |
Advanced reliability, availability, and serviceability (RAS) capabilities |
| Advanced RAS capabilities |
| - |
Intel SGX up to 128 GB max enclave size (up to 512 GB max enclave size on select SKUs) |
| Intel SGX up to 128 GB max enclave size |
| Intel SGX up to 64 GB max enclave size |
Workload acceleration with Intel QAT, Intel DLB, Intel DSA, and Intel IAA |
| Workload acceleration with Intel QAT, Intel DLB, Intel DSA, and Intel IAA |
| Workload acceleration with Intel QAT, Intel DLB, Intel DSA, and Intel IAA |