Business News

Key Announcement Made By Intel At International Supercomputing Conference

At the recently held (24 June - 2 July) International Supercomputing Conference (ISC), Intel demonstrated how the company seeks to expand its lead in the high-performance computing (HPC) space with a series of technology disclosures, partnerships, and customer acceptance.

By contentwriteramisha

Key Announcement Made By Intel At International Supercomputing Conference

The chip giant announced advancements in its Xeon processors for HPC and AI, as well as innovations in memory, software, exascale-class memory, and networking technologies for a range of HPC use cases.

"To maximize HPC's performance, we must leverage all of our computing resources and advances in technology," said Trish Damkroger, vice president and general manager of high-performance computing at Intel. “Intel is the driving force behind the industry’s move towards exascale computing, and the progress we are making with our CPUs, XPUs, OneAPI toolkit, exascale-class DAOS storage, and high-speed networking puts us closer to that realization.” 

Earlier this year, Intel strengthened its position in the HPC arena with the introduction of the third-generation Intel Xeon Scalable processors. The latest processor delivers up to 53% higher performance on a range of HPC workloads, including life sciences, financial services, and manufacturing, compared to previous generation processors.

Compared to its closest x86 competitor, the scalable third-generation Intel Xeon processors are said to deliver superior performance on a range of popular HPC workloads. The Xeon Scalable 8380 processor is also said to outperform the AMD EPYC 20 EPYC 7763 processor for critical AI workloads with 50% better performance on popular benchmarks.

HPC Labs, supercomputing centers, universities, and original equipment manufacturers that have adopted Intel's latest computing platforms are Dell Technologies, HPE, the Korea Meteorological Administration, Lenovo, the Max Planck Computing, and Data Facility, Oracle, Osaka University, and the University of Tokyo.

The company claims that its next-generation scalable Xeon processors (codenamed "Sapphire Rapids") will feature built-in High Bandwidth Memory (HBM) that delivers a dramatic increase in memory bandwidth and significant performance improvements for HPC applications that operate on memory bandwidth. Users can handle workloads only with high bandwidth memory or in combination with DDR5.
The first successes for the Sapphire Rapids processor with integrated HBM include the US Department of Energy's Aurora supercomputer at Argonne National Laboratory and the Crossroads supercomputer at Los Alamos National Laboratory.

The platform based on Sapphire Rapids provides features to accelerate HPC, including increased I/O bandwidth with PCI Express 5.0 (compared to PCI Express 4.0) and Compute Express Link (CXL) 1.1 support, which enables computing, networking and enables extended use cases across areas of storage.

In addition to memory and I/O improvements, the Sapphire Rapids is optimized for HPC and artificial intelligence (AI) workloads with a new integrated AI acceleration engine called Intel Advanced Matrix Extensions (AMX).

Intel Ethernet Extended for HPC

At ISC 2021, Intel also announced its new High-Performance Networking with Ethernet (HPN) solution, which enables small groups in the HPC segment using standard Intel Ethernet 800 series network adapters and controllers, switches based on the Intel Tofino P4 programmable. Extends capabilities to Ethernet technology. Ethernet Switch ASIC and Intel Ethernet Fabric Suite software. HPN is designed to provide application performance comparable to InfiniBand at a lower cost while taking advantage of the ease of use of Ethernet.
Commercial support for DAOS.



Intel has introduced commercial support for DAOS (Distributed Asynchronous Object Storage), an open-source software-defined object storage system designed to optimize data exchange in the Intel HPC architecture. DAOS forms the basis of the Intel Exascale memory stack previously announced by Argonne National Laboratory and is used by Intel customers such as LRZ and JINR (Joint Institute for Nuclear Research).
DAOS support is now available to Partners as an L3 support offering that enables Partners to provide a complete turnkey storage solution in conjunction with their services. In addition to Intel's own data center modules, early partners in this new commercial support include HPE, Lenovo, Supermicro, BrightSky, Croit, Netrix, Quanta, and RSC Group.

.

What's Your Reaction?

like
0
dislike
0
love
0
funny
0
angry
0
sad
0
wow
0