While Intel’s grip on the processor market is not as iron-tight as it once was, the company is still the principal player. To avoid relinquishing further ground to rivals – most notably AMD – Intel is going all-in on power. With its latest CPU and GPU offerings, the company is using high-performance processing to fend off its rivals with the new Intel Xeon Max series.
Xeon Mac CPU
Starting with the CPU, the Xeon Max supports DDR5 memory PCIe 5.0, and Compute Express Link (CXL) 1.1. There are 56 performance cores taken from the Golden Cove architecture Intel is already using in its 12th-Gen Core processors.
Furthermore, the Xeon Max features 20 accelerators to help with High-Performance Computing (HPC) workloads and also AI/machine learning:
- Intel Deep Learning Boost (DL Boost)
- Intel Advanced Vector Extensions 512 (AVX-512)
- Intel Data Streaming Accelerator (DSA)
- Intel Advanced Matrix Extensions (AMX)
Intel points to the following benefits of the Xeon Max CPU:
- “68% less power usage than an AMD Milan-X cluster for the same HPCG performance.
- AMX extensions boost AI performance and deliver 8x peak throughput over AVX-512 for INT8 with INT32 accumulation operations.2
- Provides flexibility to run in different HBM and DDR memory configurations.
- Workload benchmarks:
- Climate modeling: 2.4x faster than AMD Milan-X on MPAS-A using only HBM.
- Molecular dynamics: On DeePMD, 2.8x performance improvement against competing products with DDR memory.”
Xeon Max GPU
Next is the Xeon Max GPU, which is targeting Nvidia as the leading name in datacenter GPUs. Intel describes the package as the “highest density processor” with over 100 billion transistors on a system of 47 chiplets.
It features 128 cores built on Intel’s Xe HPC microarchitecture. There are also 128 ray tracing units that are specifically in place for simulation software and for generating digital content. Below are the benefits Intel says the Xeon Max GPU brings:
- “408MB of L2 cache – the highest in the industry – and 64MB of L1 cache to increase throughput and performance.
- The only HPC/AI GPU with native ray tracing acceleration, designed to speed scientific visualization and animation.
- Workload benchmarks:
- Finance: 2.4x performance gain over NVIDIA’s A100 on Riskfuel credit option pricing.
- Physics: 1.5x improvement over A100 for NekRS virtual reactor simulations.”
Tip of the day: Did you know you can use Windowss built in antivirus Microsoft Defender also with scheduled scans? In our tutorial we give you step-by-step instructions on how to program your personal scan-schedule to keep your free of malware.