Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


GigaIO and Microchip Power Native PCI Express Network Fabric for Composable Disaggregated Infrastructure

Carlsbad, CA – January 17, 2021 – GigaIO, maker of data center network architecture and connectivity solutions, has announced a collaboration with Microchip Technology Inc. to power GigaIO’s FabreX, the native PCI Express (PCIe) Gen4 network fabric, which supports GDR, MPI, TCP/IP and NVMe-oF. FabreX technology revolutionizes rack-scale architectures, enabling software-defined, dynamically reconfigurable rack-scale systems, […]

HPC Predictions 2021: Quantum Beyond Qubits, Rising Composable and Memory-Centric Tide, Neural Network Explosion and 5G’s Slow March

Annual technology predictions are like years: some are better than others. Nonetheless, many provide useful insight and serve as the basis for worthwhile discussion. Over the last few months we received a number of HPC and AI predictions for 2021, here are the most interesting and potentially valid. Let’s check in 12 months from now […]

At SC20: GigaIO Launches PCIe 4.0 Accelerator Pooling Appliance for the Composable Data Center

CARLSBAD, CA – November 17, 2020 – GigaIO Networks, creators of data center network architecture and connectivity solutions, today announced their new Hydra product line, which the company said is the industry’s first fully managed PCI Express (PCIe) Gen 4.0 Pooling Appliance, a high-performance expansion chassis for the disaggregating and pooling of PCIe accelerator devices. […]

Intelligent Fabrics for the Next Wave of AI Innovation

In this sponsored post, our friend John Spiers, Chief Strategy Officer at Liqid, discusses how resource utilization and the soaring costs surrounding it are a constant push and pull issue for IT Departments. Now with the emergence of AI and machine learning, resource utilization is far more front and center than it has ever been. Managing legacy hardware in a hyperconverged environment just as you always have is not going to cut it, because the people and hardware costs associated with these extremely heavy workloads are tremendous. Intelligent fabrics and composable infrastructure software deliver a solution to the problem that allows IT providers the ability to pool and deploy their hardware resources to adapt to the workload at hand, then re-deploy as required for a balanced system that can address the demands of AI and machine learning.

Video: GigaIO on Optimizing Compute Resources for ML, HPDA and other Advanced Workloads

In this interview, GigaIO CEO Alan Benjamin talks about systems performance problems and wasted compute resources when implementing ML, HPDA and other high demand workloads that involve high data volumes. At issue, Benjamin explains, is today’s rack architecture, which is decades old and unsuited for combinations of CPUs, GPUs and other accelerators needed for advanced computing strategies. The answer: the “composable disaggregated infrastructure.”

Composable Computing at SDSC

In this Q&A, SDSC Chief Data Science Officer Ilkay Altintas explains the rationale for composable systems and the approach taken with the new Expanse supercomputer. With its new Expanse supercomputer, San Diego Supercomputer Center (SDSC) is pioneering composable HPC systems to enable the dynamic allocation of resources tailored to individual workloads. One of the critical innovations in the SDSC’s new Expanse supercomputer from Dell Technologies, is the ability to support composable systems with dynamic capabilities.

vScaler Integrates SLURM with GigaIO FabreX for Elastic HPC Cloud Device Scaling

Open source private HPC cloud specialist vScaler today announced the integration of SLURM workload manager with GigaIO’s FabreX for elastic scaling of PCI devices and HPC disaggregation. FabreX, which GigaIO describes as the “first in-memory network,” supports vScaler’s private cloud appliances for such workloads such as deep learning, biotechnology and big data analytics. vScaler’s disaggregated […]

LIqid Steps up with Composable Infrastructure at SC19

In this video from SC19, Sumit Puri from Liqid describes the company’s innovative composable infrastructure technology for HPC. “We don’t build servers statically. We build servers dynamically by taking software and reconfiguring servers on the fly to have any amount of storage, GPU, networking, or compute that the application layer requires. Our mission is to turn the data center from statically configured to dynamically configurable.”

Liqid Enables Multi-Fabric Support for Composable Infrastructure

Today Liqid announced unified multi-fabric support for composability across all major fabric types including PCIe Gen 3, PCIe Gen 4, Ethernet, Infiniband, and laying the foundation for the up-coming Gen-Z specifications. “Providing Ethernet and Infiniband composability in addition to PCIe is a natural extension of our expertise in fabric management and aligns with our mission to facilitate data center disaggregation,” said Sumit Puri, CEO and Co-founder, Liqid.

One Stop Systems Showcases Composable Infrastructure for GPU Workloads at ISC 2018

In this video from ISC 2018, Jaan Mannik from One Stop Systems describes the company’s HPC systems and new composable infrastructure solutions. OneStop also showcased a wide array of its high-density NVIDIA GPU-based appliances, as well as showcase a live remote connection to one of its machine learning and HPC platforms. “OSS leads the market in external systems that increase a server’s performance in HPC applications, reducing cost and impact on data center infrastructure. These technology-hungry applications include AI (artificial intelligence), deep learning, seismic exploration, financial modeling, media and entertainment, security and defense.”