Skip to main content
Intel® Accelerator Engines

Redefine performance with 4th Gen Intel® Xeon® Scalable processors – featuring built-in accelerators to improve performance across the fastest-growing workloads in AI, data analytics, networking, storage, and HPC.

Photo of a 4th Gen Intel® Xeon® Scalable processor
General Compute

53% average performance gain over the prior generation1

Artificial Intelligence

up to 10x higher PyTorch real-time inference performance with built-in Intel® (BF16) vs. the prior generation (FP32)2

HPC

up to 56% higher average HPC performance vs. the prior generation3

Networking

up to 47% fewer cores to achieve the same connections/second using integrated Intel® QAT vs. the prior generation on NGINX key handshake4

5G

up to 2x the capacity at the same power envelope for vRAN workloads vs. prior generation processors5

Intel® IAA

up to 3x higher RocksDB performance using integrated Intel® IAA vs. the prior generation6

Intel® DSA

up to 60% higher IOPS and up to 37% latency reduction for large packet sequential read using integrated Intel® DSA vs. the prior generation7

  1. See [G1] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  2. See [A17] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  3. See [H1] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  4. See [N15] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  5. See [N10] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  6. See [D1] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  7. See [N18] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
Supermicro X13 Workload-Optimized Systems

AI

Intel AMX

up to 10x higher PyTorch real-time inference performance

Intel AMX

up to 10x higher PyTorch training performance

(BF16) vs. the prior generation (FP32)1

Deep Learning Training and Inference

Run AI directly on the CPU without the need for discrete GPU cards. Supermicro X13 systems support high-power CPUs with liquid cooling options, maximum memory footprint and latest generation NVMe storage. Optimized Supermicro X13 systems include Hyper, BigTwin®, GrandTwin and CloudDC.

Use Cases: Natural Language Processing, Recommender Systems

Intel® Advanced Matrix Extensions (Intel® AMX)

Significantly accelerate AI capabilities on the CPU with Intel® AMX. Intel AMX is a built-in accelerator that improves the performance of deep learning training and inference on 4th Gen Intel® Xeon® Scalable processors, ideal for workloads like natural language processing, recommendation systems, and image recognition.

  1. See [A17] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
Supermicro X13 2U Hyper system SYS-221H-TNR
X13 Hyper – SYS-221H-TNR

Analytics

Intel IAA

up to 3x higher RocksDB performance vs.the prior generation2

Intel IAA

up to 59% higher ClickHouse DB performance3

Analytics Unleashed

Improve speed and efficiency of CRM systems with Supermicro X13 systems powered by 4th Gen Intel Xeon Scalable Processors. Supermicro X13 GrandTwin and Hyper deliver optimal memory density for database applications, while 4-way and 8-way multi-processor systems allow up to 32TB of DDR5 in 128 slots.

Use Cases: In-memory Database, ERP, CRM

Intel® In-Memory Analytics Accelerator (Intel® IAA)

Run database and analytics workloads faster, with potentially greater power efficiency. Intel® IAA increases query throughput and decreases the memory footprint for in-memory database and big data analytics workloads. Intel IAA is ideal for in-memory databases, open-source databases, and data stores.

  1. See [D1] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  2. See [D2] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
Supermicro X13 2U 4-way Multi-Processor system SYS-241H-TNRTTP
X13 Multi-Processor – SYS-241H-TNRTTP

Storage

Intel QAT

up to 95% fewer cores required and up to 2x higher level 1 compression4

Intel DSA

up to 60% higher IOPS and up to 37% latency reduction for sequential read5

Streamlined Storage

Supermicro’s X13 storage architectures have been designed for maximum performance, compatible with the latest high-speed flash memory standards and featuring symmetrical layouts to minimize signal paths for the lowest possible latency. From maximum throughput Petascale All-Flash to scale up/out Top-Loading systems, Supermicro X13 has an optimized storage solution for any application.

Use Cases: Software-defined Storage, Data-intensive HPC, Private/Hybrid Cloud

Intel Data Streaming Accelerator (Intel® DSA)

Designed to offload the most common data movement tasks that cause overhead in data center–scale deployments. Intel DSA helps speed up data movement across the CPU, memory, and caches, as well as all attached memory, storage, and network devices.

  1. Using integrated QAT. See [N16] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  2. See [N18] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
Supermicro X13 1U Petascale All-Flash storage server SSG-121E-NE316R
X13 Petascale All-Flash – SSG-121E-NE316R

Networking

Intel QAT

up to 47% fewer cores required to achieve the same connections/second6

Intel QAT

up to 95% higher vSwitch throughput for packet sizes above 800B for bi-directional switching7

Maximize Networking Throughput

Supermicro systems are optimized for maximum efficiency in networking applications, further enhanced when switching and encryption workloads are offloaded to the built-in Intel QuickAssist Technology accelerator engine on 4th Gen Intel Xeon Scalable processors. Supermicro X13 CloudDC and WIO family rackmount servers provide an all-in-one solution for cloud data centers, combining optimal compute, memory and storage density with balanced TCO.

Use Cases: Cloud, Virtualization, HCI, Hosting, Content Delivery

Intel® QuickAssist Technology (Intel® QAT)

Help reduce system resource consumption – and TCO – by accelerating cryptography and data compression with Intel® QAT. By offloading encryption, decryption, and compression, this built-in accelerator helps free up processor cores so that systems can serve a larger number of clients or use less power. Get the most value when you merge compression and encryption – with Intel QAT, 4th Gen Intel® Xeon® Scalable processors are the highest-performance CPU that can compress and encrypt in a single data flow.

  1. See [N15] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
  2. See [W8] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
Supermicro X13 2U CloudDC System SYS-521C-NR
X13 CloudDC – SYS-521C-NR

5G

Intel vRAN Boost

up to 2x the capacity and the same power envelope for vRAN workloads8

Intel vRAN Boost

Reduce power consumption by up to 20%

Efficiency at the Edge

Supermicro is accelerating innovation at the intelligent edge with new telco-optimized systems powered by 4th Gen Intel Xeon Scalable Processors. Supermicro X13 5G systems feature compact chassis designs, front I/O and optional DC power, easily integrating into existing infrastructure and delivering data center performance and maximum efficiency in challenging edge environments.

Use Cases: vRAN, ORAN, Multi-Access Edge Computing

with Intel® vRAN Boost

Intel® vRAN Boost supercharges vRAN and Open vRAN with outstanding performance-per-watt to meet operators’ critical scaling, flexibility, and energy efficiency requirements. Its fully-integrated vRAN acceleration reduces system complexity and helps improve total cost of ownership.

  1. See [N10] at intel.com/processorclaims: 4th Gen Intel® Xeon® Scalable processors. Results may vary.
X13 SuperEdge short-depth 2U 3-node edge server SYS-211SE-31D
X13 SuperEdge – SYS-211SE-31D
Resources