Supercomputing Innovations and Breakthroughs
HPC and AI at Exascale
Intel is engineering the first US exascale supercomputer, built on future Intel® Xeon® Scalable processors, Intel® Xeon® compute architecture, Xeon® Optane™ DC Persistent Memory, and more.
Manufacturing, Life Sciences, and Oil & Gas
HPC is rapidly transforming as Intel® Xeon® Scalable processors power the next wave of analytics, simulation & modeling, and AI.
KISTI Changes Supercomputing in Korea
KISTI with its NURION supercomputer is growing to broaden the leadership of HPC R&D in Korea.
Supporting Demanding Computational Science
The Marconi supercomputer is a critical step to Cineca achieving their exascale computing objectives.
Osaka Supercomputer Supports Research
Osaka University's OCTOPUS Supercomputer has boosted scientific capacity and supports a wide range of projects.
Introducing the 2nd Generation Intel® Xeon® Scalable Processors
Discover the flexibility of one platform to handle your AI, simulation and modeling, analytics and other general HPC workloads.
Performance for HPC Platforms
Our latest platform with the 2nd Gen Intel® Xeon® Scalable processor reduces the need for dedicated systems running specialized hardware and software for unique workloads.
Leadership Compute Performance
Learn how the new Intel® Xeon® Platinum 9200 processors and Intel® Server System S9200WK are designed to optimize performance for HPC and AI workloads.
HLRN Brings Advanced Performance to HPC
HLRN selected the new Intel® Xeon® Platinum 9200 processors to meet their increasingly diverse needs for HPC workloads.
TACC Offers New HPC Capabilities
The Texas Advanced Computing Center (TACC) is using 2nd Generation Intel® Xeon® Scalable processors to solve simulation-based and data-driven science problems.
The Convergence of AI and HPC
AI on the Intel®-based HPC Infrastructure You Know
Artificial intelligence (AI) has the potential to change our lives. It’s powerful and far reaching yet does not require separate highly specialized hardware and software systems to run. Your familiar Intel®-based HPC infrastructure handles AI along with your big-data analytical workloads and more: For more insight, discovery, and competitive advantage.
Watch “AI and HPC Workload Convergence” ›
Running AI and Analytics in a Single Cluster Infrastructure
This solution brief introduces the challenges and opportunities around running AI and analytics workloads on existing HPC clusters.
AI on HPC Infrastructure in Three Steps
There are a number of steps to successfully integrating AI workloads into your existing HPC environment — our visual guide will help you get started.
Supporting HPC, AI, Analytics on a Common Platform
Learn about capabilities for running TensorFlow* and Apache Spark* on existing HPC systems using standard batch schedulers.
Intel® Select Solutions for HPC
Intel® Select Solutions are pre-configured HPC stacks designed and optimized for specific data-center workloads — like professional visualization, simulation and modeling, and genomic analytics. Available through Intel’s trusted OEM partners, Intel® Select Solutions will accelerate your HPC plans.
Intel® HPC Platform Specification and Community
Intel® HPC Platform Specification
See the recommended combination of compute, fabric, memory, storage, and software elements that comprise the fundamental requirements for Intel® Select Solutions.
Intel® HPC Application Catalog
View a centralized list of applications verified compatible with Intel® HPC Platform Specification and Intel® Select Solutions for HPC.
Intel® Cluster Checker Tool
Verify that cluster components work together seamlessly for improved uptime, productivity, and lower total cost of ownership (TCO).
Intel HPC Products and Technology
Intel® Xeon® Scalable Processors
Drive actionable insight, count on hardware-based security, and deploy dynamic service delivery with Intel® Xeon® Scalable processors.
Intel® Deep Learning Boost
Intel® Xeon® Scalable processors take embedded AI performance to the next level with Intel® Deep Learning Boost (Intel DL Boost).
Intel® Optane™ Technology
We’re enabling solutions that unleash CPU utilization, reduce bottlenecks, and deliver unprecedented insights from large datasets.
Intel® Omni-Path Architecture
Intel® Omni-Path Architecture (Intel® OPA) lowers system TCO while providing reliability, high performance, and extreme scalability.
Data Center Storage Solutions
With the explosion of data, modernizing storage is critical to IT transformation. Advances in technology allow for more efficient storage, access, and transfer of data.
Intel® FPGAs
From the IoT to the data center, Intel® FPGA solutions deliver the speed and capacity of full systems-on-chip.
Intel® Ethernet Network Adapter
Intel® Ethernet Network Adapters, Controllers, and Accessories deliver services efficiently and cost-effectively in the data center.
HPC Software and Tools
Modernize your code for today's and tomorrow’s hardware using advanced tools that help build, debug, and tune your applications.
Never Miss Out on the Most Important News
Get Intel insight behind each week's technology headlines. Sample a bit of Intel's perspective on this week's developments and then subscribe today.
By signing up, you agree to our Privacy and Terms of Use.
By signing up, you agree to our Privacy and Terms of Use.
Informacje o produktach i wydajności
30-krotny wzrost przepustowości wnioskowania w przypadku procesora Intel® Xeon® Platinum 9282 z technologią Intel® Deep Learning Boost (Intel DL Boost): testy przeprowadzone przez firmę Intel 26 lutego 2019 roku. Platforma: Dragon Rock, 2-gniazdowy procesor Intel® Xeon® Platinum 9282 (56 rdzeni na gniazdo), HT WŁ., technologia Turbo WŁ., pamięć całkowita: 768 GB (24 gniazda / 32 GB / 2933 MHz), system BIOS: SE5C620.86B.0D.01.0241.112020180249, Centos* 7, jądro 3.10.0-957.5.1.el7.x86_64, Deep Learning Framework: optymalizacja Intel® dla architektury Caffe*, wersja: https://github.com/intel/caffe d554cbf1, ICC 2019.2.187, wersja MKL DNN: v0.17 (commit hash: 830a10059a018cd2634d94195140cf2d8790a75a), model: https://github.com/intel/caffe/blob/master/models/intel_optimized_models/int8/resnet50_int8_full_conv.prototxt, BS=64, brak warstwy danych syntheticData: 3 × 224 × 224, 56 instancji / 2 gniazda, typ danych: INT8 w porównaniu do testów firmy Intel z 11 lipca 2017 r.: 2-gniazdowy procesor Intel® Xeon® Platinum 8180, 2,50 GHz (28 rdzeni), HT wył., technologia Turbo wył., mechanizm zarządzania skalowaniem ustawiony na „performance” w sterowniku intel_pstate, 384 GB pamięci RAM DDR4-2666 z funkcją ECC. CentOS* Linux, wersja 7.3.1611 (Core), jądro systemu Linux 3.10.0-514.10.2.el7.x86_64. Dysk SSD: Intel® SSD z serii DC S3700 (800 GB, 2,5 cala SATA 6 Gb/s, 25 nm, MLC). Wydajność zmierzono przy ustawieniach: Zmienne środowiskowe: KMP_AFFINITY='granularity=fine, compact‘, OMP_NUM_THREADS=56, CPU Freq set with cpupower frequency-set -d 2.5G -u 3.8G -g performance. Caffe: (http://github.com/intel/caffe/), revision f96b759f71b2281835f690af267158b82b150b5c. Wyciąganie wniosków zmierzono poleceniem „caffe time --forward_only”, szkolenie zmierzono poleceniem „caffe time”. W przypadku topologii „ConvNet” użyto syntetycznego zbioru danych. W przypadku pozostałych topologii dane przechowywano w lokalnej pamięci masowej i buforowano w pamięci operacyjnej przed szkoleniem. Dane techniczne topologii: https://github.com/intel/caffe/tree/master/models/intel_optimized_models (ResNet-50). Kompilator Intel® C++, wer. 17.0.2 20170213, małe biblioteki Intel® Math Kernel Library (Intel® MKL), wersja 2018.0.20170425. Caffe uruchomiono z parametrem „numactl -l”.
Czterokrotnie wyższa wydajność w testach Linpack w przypadku procesora Intel® Xeon® Platinum 9242 drugiej generacji w porównaniu z procesorem AMD* EPYC* 7601 ze skalowaniem (4-węzłowym, 8-węzłowym).
Procesor Intel® Xeon® 9242: platforma referencyjna firmy Intel z procesorami Intel® Xeon® 9242 2S (2,2 GHz, 48 rdzeni), 16 × 16 GB pamięci DDR4-2933, 1 dysk SSD, system plików klastra: 2.12.0-1 (serwer) 2.11.0-14.1 (klient), BIOS: PLYXCRB1.86B.0572.D02.1901180818, mikrokod: 0x4000017, CentOS* 7.6, jądro: 3.10.0-957.5.1.el7.x86_64, stos OFED: OFED OPA 10.8 na RH7.5 z Lustre v2.10.4, adapter magistrali hosta: architektura Intel® Omni-Path Architecture (Intel® OPA), jednoportowy adapter PCIe* × 16, 100 Gb/s, przełącznik: (Intel® OPA) Edge Switch z serii 100, 48 portów, HPL 2.1, kompilator Intel 2019u1, biblioteka Intel® Math Kernel Library (Intel® MKL) 2019, Intel MPI 2019u1, HT = wł., Turbo = wył., 2 wątki na rdzeń, 4-węzłowy = 20 408,00, 8-węzłowy = 39921 GF/s, im wyższa wartość, tym lepiej, test firmy Intel 03.03.2019 r.
AMD EPYC 7601: Supermicro AS -1023US-TR4, AMD EPYC 7601 (2,2 GHz, 32 rdzenie) 2S, 16 × 16 GB pamięci DDR4-2666, 1 dysk SSD, wersja systemu BIOS: 1.1b (20.08.2018), mikrokod: 0x8001227, Oracle* Linux Server, wer. 7.5 (3.10.0-862.14.4.el7.crt1.x86_64), system plików klastra: Panasas (pamięć masowa 124 TB) wersja oprogramowania sprzętowego 5.5.0.b-1067797.15 EDR based IEEL Lustre, Mellanox EDR MT27700, 100Gb/s, 36-portowy przełącznik Mellanox EDR IB Switch, OFED MLNX mlnx-4.3-3.0.2.0, HPL 2.2, kompilator Intel 2018u3, AMD BLIS v0.4.0, Intel MPI 2018u3, SMT = wł., Turbo = wł., 2 wątki na rdzeń, 4-węzłowy = 4739,96, 8-węzłowy = 9406,07 GF/s, im wyższa wartość, tym lepiej, test firmy Intel 23.09.2018 r.