Unlocking the Potential of High-Performance Computing

Today, we launched the new Intel® Xeon® Scalable platform that will fuel the next generation of personalized medicine, product development, energy exploration, weather modeling, and numerous other high-performance computing (HPC) applications. Across virtually all industries, HPC systems give engineers and researchers the tools they need to drive innovation and discovery to solve the big challenges — problems that may have seemed all but unsolvable on yesterday’s systems. The use of HPC to gain actionable insights is fundamental.

This march forward to tackle ever-larger problems involves harnessing ever-larger data sets and creates the need for new generations of HPC systems. Today’s HPC users need systems with innovative technologies in compute, memory, fabric, storage, and optimized software that provides balanced, scalable performance and breaks system bottlenecks to unleash the next generation of innovation.

That’s the idea behind the new Intel Xeon Scalable platform, designed to deliver advanced HPC capabilities. Data scientists, engineers, and researchers looking to gain faster insights from data and accelerate product innovation will benefit from enhanced capabilities across compute, storage, memory, and I/O.

This platform represents a significant leap forward in the performance and efficiency of cutting-edge HPC systems. An innovative approach to platform design in the Intel Xeon Scalable platform unlocks performance for a broad range of HPC systems — from the smallest clusters all the way to the world’s largest supercomputers.

This new processor family delivers leading performance across a broad portfolio of balanced platforms. The Intel Xeon Scalable platform yields an increase of up to 8.2x more double precision GFLOPS/sec when compared to Intel® Xeon® processor E5 (formerly codenamed Sandy Bridge), and 2.27x increase over the previous-generation Intel Xeon processor E5 v4 (formerly codenamed Broadwell)[1]. With up to 28 cores connected through Intel® Mesh architecture, as well as significant increases in memory and I/O bandwidth (with six memory channels and 48 PCIe* lanes), we’re talking about a wide-open freeway for extremely large compute-and-data-intensive workloads.

In another gain for HPC users, the Intel Xeon processor Scalable family incorporates a new generation of Intel® Advanced Vector Extensions called Intel AVX-512. With up to double the flops per clock cycle compared to the previous-generation Intel® AVX2, the new Intel® AVX-512 boosts performance for demanding computational workloads in applications such as modeling and simulation, data analytics, machine learning, and visualization.

In yet another notable step forward, the new Intel Xeon Scalable platform includes processors that integrate Intel® Omni-Path Architecture (Intel® OPA). Intel OPA provides 100Gbps bandwidth with low latency for HPC clusters. With this integration, users can take advantage of a fabric built for HPC without consuming I/O ports and slots, freeing those up for other platform uses.

Beyond the processor, the Intel Xeon Scalable platform supports next-generation PCIe-based Intel® Optane™ solid state drives. Built with revolutionary 3D XPoint™ memory media, these SSDs are the first products to combine the attributes of memory and storage. With an industry-leading combination of high throughput, low latency, high quality of service, and ultra-high endurance, this innovative storage solution is optimized to break through data access bottlenecks by providing a new data storage tier for your HPC workloads.

Finally, Intel offers tools and libraries to help developers optimize HPC software capabilities and performance. To help HPC users take full advantage of system resources, Intel offers Intel® HPC Orchestrator system software that simplifies the development, installation, and ongoing maintenance of the HPC system software stack. To help developers get the full value from an Intel based solution, Intel delivers optimizations for popular AI deep learning frameworks including neon™, Caffe*, Theano*, Torch* and TensorFlow* as well as performance libraries like Intel® Math Kernel Library for Deep Neural Networks (Intel® MKL-DNN) to accelerate deep learning inference and training on Intel architecture and Intel® Data Analytics Acceleration Library (Intel® DAAL) to speed big data analytics. These contributions to the community increase the performance and value of HPC systems for researchers and engineers working on Intel-based platforms.

The new Intel Xeon Scalable platform equips scientists and engineers with a powerful computational tool, an enabler and accelerant for the work done by bright, creative people who are focused on solving the world’s biggest problems. Our world needs safer forms of transportation, new treatments for cancer, more efficient energy sources, smarter cities, more intelligent security systems, and much more. To solve these problems — these very big problems — we need to unleash the potential of the HPC community.

That’s really the story behind the new Intel Xeon Scalable platform: empowering the HPC community to rethink the impossible and make the world a better place. We could not be more excited to help.

To learn more about the new Intel Xeon Scalable platform, visit intel.com/hpc or contact your preferred system vendor.

 

[1]Baseline config: 1-Node, 2 x Intel® Xeon® Processor E5-2699 v4 on Red Hat Enterprise Linux* 7.0 kernel 3.10.0-123 using Intel® Distribution for LINPACK Benchmark, score: 1446.4 GFLOPS/s vs. estimates based on Intel internal testing on 1-Node, 2x Intel Xeon Scalable processor (codename Skylake-SP) system. Score: 3295.57

Baseline config:  1-Node, 2 x Intel® Xeon® Processor E5-2690 based system on Red Hat Enterprise Linux* 6.0 kernel version 2.6.32-504.el6.x86_64 using Intel® Distribution for LINPACK Benchmark. Score: 366.0 GFLOPS/s vs. 1-Node, 2 x Intel® Xeon® Scalable processor on Ubuntu 17.04 using MKL 2017 Update 2. Score: 3007.8

Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products.  For more complete information visit  www.intel.com/benchmarks.

Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at [intel.com].

 

Published on Categories High Performance ComputingTags , , ,
Trish Damkroger

About Trish Damkroger

Trish Damkroger is Vice President of Intel’s Data Center Group and General Manager of its Technical Computing Initiative. Her work helps shape Intel’s high-performance computing (HPC) products and services for the technical market segment. Under this umbrella are the next-generation platform technologies and frameworks that will take Intel toward exascale and advance the convergence of traditional HPC, big data, and artificial intelligence workloads. Damkroger has more than 27 years of experience in technical and managerial roles both in the private sector and during her 15-year career within the United States Department of Energy. While at the Department of Energy, she served most recently as the Associate Director of Computation (Acting) at Lawrence Livermore National Laboratory leading a 1,000 person group that is one of the world’s leading teams of supercomputing and scientific experts. Since 2006, Damkroger has been a leader of the annual Supercomputing Conference series, the premier international meeting for high performance computing. She was the SC14 General Chair, headed the SC15 steering committee, led the SC16 Diverse HPC Workforce Committee, and has signed on as Vice Chair of SC18. Damkroger is also a certified coach and a strong advocate for women in science, technology, engineering, and math (STEM). Damkroger was named one of HPCwire’s People to Watch in 2014. She has a master’s degree in electrical engineering from Stanford University.