Sign up for our newsletter and get the latest big data news and analysis.

SC22: CXL3.0, the Future of HPC Interconnects and Frontier vs. Fugaku

HPC luminary Jack Dongarra’s fascinating comments at SC22 on the low efficiency of leadership-class supercomputers highlighted by the latest High Performance Conjugate Gradients (HPCG) benchmark results will, I believe, influence the next generation of supercomputer architectures to optimize for sparse matrix computations. The upcoming technology that will help address this problem is CXL. Next generation architectures will use CXL3.0 switches to connect processing nodes, pooled memory and I/O resources into very large, coherent fabrics within a rack, and use Ethernet between racks. I call this a “Petalith” architecture (explanation below), and I think CXL will play a significant and growing role in shaping this emerging development in the high performance interconnect space.

SC22: Jeff McVeigh on Intel’s 3 New Server Chips; oneAPI and Heterogeneity; and an Aurora Update

We caught up with Jeff McVeigh, Corporate Vice President and General Manager of Intel’s Super Compute Group, to learn more about the three new server chips Intel recently announced, the Max CPU and the Max GPU series, along with the 4th Generation Xeon Scalable Processor. We also discussed Intel’s oneAPI and how it supports heterogeneous HPC […]

At SC22: DDN’s Kurt Kuckein Talks Advanced AI Data Management at Scale

At SC22, we caught up with DDN Senior Vice President of Marketing to discuss major trends driving AI data management at scale – trends that include growing adoption of GPUs for AI workloads, the emergence of other accelerator technology, DDN products and services in support of fast integration of these new technologies, along with growing […]

Tokyo University of Science: Research Using Data Science, Materials Informatics May Lead to Better EV Motors

Dec. 1, 2022 — Tokyo University of Science today said researchers have combined several disciplines to investigate coercity, which is the physical property of magnetic materials that has the potential to advance electric car motor efficiency. The researchers joined together data science, materials informatics,and an extension of the Ginzburg–Landau model to explain how coercivity – physical […]

Oak Ridge Lab and Veterans Affairs Use Summit Supercomputer Security Framework for Health Research

Dec. 1, 2022 — A team from Oak Ridge National Laboratory (ORNL) and the US Department of Veterans Affairs (VA) used the CITADEL security framework to securely transfer and analyze veterans’ health records on ORNL’s Summit, an IBM AC922 supercomputer housed at the Oak Ridge Leadership Computing Facility, a US Department of Energy (DOE) Office […]

Argonne Researchers Win Gordon Bell Special Prize for Using Language Models to Track Virus Variants

Nov. 29, 2022 — Scientists from the U.S. Department of Energy’s (DOE) Argonne National Laboratory and a team of collaborators have won the Association for Computing Machinery’s 2022 Gordon Bell Special Prize for High Performance Computing-Based COVID-19 Research for their new method of quickly identifying how a virus evolves. Their work in training large language models (LLMs) […]

AWS Announces Hpc7g EC2 Instance Powered by New Arm-based Graviton3E Chip

At AWS’s re:Invent conference yesterday, Amazon Web Servicesannounced three EC2 instances powered by new Arm-based chips designed by AWS. AWS said its Hpc7g instances on EC2, powered by new AWS Graviton3E chips, offer up to 2x better floating-point performance compared to current generation C6gn instances and up to 20 percent higher performance compared to current […]

Exxact Partners Offers Run:ai for GPU Clusters in AI Workloads

FREMONT, CA — Nov. 30, 2022 — Exxact Corporation, a provider of high-performance computing (HPC), artificial intelligence (AI), and data center solutions, now offers Run:ai in their solutions. This groundbreaking Kubernetes-based orchestration tool incorporates an AI-dedicated, high-performant super-scheduler tailored for managing GPU resources in AI clusters. Run:ai dynamically optimizes hardware utilization for AI workloads, enabling clusters […]

Jan. 20 Deadline for Abstracts for Energy HPC Conference at Ken Kennedy Institute, Rice Univ.

Nov. 30, 2022 — Friday, Jan. 20, 2023 is the deadline to submit an extended abstract for presentation at the 16th Annual 2023 Energy High Performance Computing Conference hosted by the Ken Kennedy Institute at Rice University, February 28 – March 2, 2023. ​Acceptance notices will be issues on February 1. ​The conference invites prospective speakers […]

Iceotope Study with Meta on Immersion Liquid Cooling for High-Density Storage Drives

Sheffield, UK, November 30 2022 — Iceotope, a precision immersion cooling company, announced a new study with Meta that Iceotope said confirms the practicality, efficiency and effectiveness of chassis-level liquid cooling technology to meet the cooling requirements of high-density storage disks increasingly being deployed and utilised by hyperscale data centre service providers. The recently published […]