New Intel Xeon Scalable Processors Accelerate HPC Systems

Print Friendly, PDF & Email

The following guest article from Intel outlines the highlights and features of the company’s new Intel Xeon scalable processors designed to accelerate HPC systems. 

The Intel Xeon processor Scalable Family, the newest Intel Xeon processors (formerly code name Skylake) are optimized to address today’s most demanding high-performance computing challenges. Part of Intel Scalable System Framework, Intel’s balanced technology foundation for HPC, the latest generation Intel Xeon processors include advanced, integrated features to help organizations speed workloads, delivering data and insights faster than ever before possible. Dramatically exceeding their predecessors’ capability, the new processors will accelerate HPC-related discoveries and innovations including utilizing emerging approaches like artificial intelligence (AI) and machine learning.

HPC Everywhere

Today, Intel Xeon processors support 97 percent of all AI workloads1. Functionality built into the processors is ideal for deploying full-scale AI inference engines, and for integration with other technical and enterprise-level applications. With more cores, more memory bandwidth, integrated fabric controllers, and a range of optional, workload-specific accelerators, Intel Xeon processors provide significant advances in performance, scalability, and flexibility.

In addition to AI innovation, the future-ready Intel Xeon processor Scalable Family’s leap forward in HPC performance benefits many industries and research institutions. Organizations tackling compute-intense workloads need faster system performance to meet the ever-increasing demands and complexity of today’s and tomorrow’s tasks. The new flagship Intel Xeon processors offer many enhancements to encourage emerging discoveries and solve large-scale and data-intense computing challenges.

In addition to AI innovation, the future-ready Intel Xeon processor Scalable Family’s leap forward in HPC performance benefits many industries and research institutions.

Power Where You Need It

Today, Top500.org’s lists, ranking the world’s fastest supercomputers, illustrates the degree to which organizations depend on Intel processors for ultimate performance. Intel Xeon processors are at the heart of more than 90 percent of today’s TOP500 systems2. The new generation of processors will propel many of these systems even further, delivering performance never before possible. With up to 28 cores onboard for compute and data-intensive workloads, plus significant increases in memory and I/O bandwidth, six memory channels and 48 PCIe lanes also contribute to enhanced throughput.

Intel Xeon processors integrate the new Intel® Advanced Vector Extensions 512 (Intel AVX 512), up to doubling the number of flops per clock cycle in comparison with the previous-generation Intel AVX2. Intel AVX 512 supports extremely demanding computational workloads like data analytics, machine learning, and visualization. The new processors also integrate Intel QuickAssist Technology, offering supplemental hardware acceleration for demanding tasks like cryptography. In addition to these features, Intel Xeon processors continue Intel’s commitment to support for common x86 instructions, cross-platform code portability, and standards-based programmability.

An optional integrated feature on the processors, Intel Omni-Path Architecture (Intel OPA), delivers 100Gbps port bandwidth and a low-latency next-generation fabric ideal for HPC systems of all sizes, even in scenarios scaling into tens of thousands of nodes. With Intel OPA directly integrated with the processor as opposed to connecting via an add-on card plugged into a PCI Express* slot, systems can be more cost-effective, plus have users more I/O flexibility by opening up these slots for other uses.

With all these enhanced features on tap, the latest Intel Xeon processors represent Intel’s best general-purpose CPU, offering great parallel and serial performance. Built from the ground up to address the critical challenges of HPC, enterprise and cloud workloads, the processors excel for compute-bound applications demanding faster cores with a larger cache.

[clickToTweet tweet=”Intel Xeon processors represent Intel’s best general-purpose CPU. #HPC” quote=”Intel Xeon processors represent Intel’s best general-purpose CPU. #HPC”]

Put Intel Xeon Scalable Processors to Work For You

Today’s business environment is highly competitive. More powerful HPC systems enable organizations to conceptualize, design, test, and deliver products to market faster than ever before. Similarly, supercomputers give researchers the ability to rapidly analyze, gain insights, and provide breakthrough discoveries from incredibly complex data sets. However, extreme-scale problems are not the only ones solved with HPC. Companies and institutions of all sizes realize the benefits and advantages HPC provides them now.

Intel’s Trish Damkroger discusses how the latest Intel® Xeon® Scalable processors, with integrated Intel® Omni-Path Architecture and Intel® AVX-512, bring significant performance leaps to the High Performance Computing community. Watch the video

Find out how Intel Xeon Scalable processors can help your organization.

______________

1 The Intel Xeon Processor Scalable Family 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]

Disclaimer:  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 process 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.

2 Intel internal estimate