Press Releases

GigaIO’s memory fabric, FabreX, to create the world’s first petabyte-scale GPU memory pool.   San Jose, California, December 6, 2023 – GigaIO, the award-winning provider of open workload-defined infrastructure for AI and accelerated computing, today announced the largest order yet for its flagship SuperNODETM, which will utilize tens of thousands of the AMD Instinct MI300X […]

GigaIO technology was deemed award-winning in September, and doubled performance by SC23. Denver, Colorado, November 15, 2023 – GigaIO, the leading provider of workload-defined infrastructure for AI and accelerated computing, is proud to announce that its groundbreaking SuperNODETM — the world’s first 32 GPU single-node supercomputer — won two coveted HPCwire Editors’ Choice Awards at […]

GigaIO technology shatters limitations for Gen AI and High Performance Computing (HPC). San Diego, California, November 9, 2023 – GigaIO, the leading provider of workload-defined infrastructure for AI and accelerated computing, today announced the launch of the world’s first 64 GPU single-node supercomputer, called SuperDuperNODETM.  GigaIO introduced a 32-GPU SuperNODE at ISC in Hamburg earlier […]

GigaIO solutions already solve obstacles predicted by Gartner to slow market adoption for other vendors. San Diego, California, August 24, 2023 – GigaIO, the leading provider of workload-defined infrastructure for AI and accelerated computing, today announced its inclusion under “Composable Infrastructure” in four of Gartner’s new reports: Hype Cycle for Infrastructure Strategy, 2023, Hype Cycle […]

GigaIO SuperNODE™ with FabreX™ dynamic memory fabric delivers unprecedented compute capabilities with AMD Instinct™ accelerators for low-power, accelerated computing. Carlsbad, California, July 13, 2023 – GigaIO, the leading provider of workload-defined infrastructure for AI and technical computing workflows, recently announced that it successfully configured 32 AMD Instinct MI210 accelerators to a single server utilizing the […]

San Diego, California, January 19, 2023 – GigaIO, the leading provider of workload-defined infrastructure without compromise for HPC + AI workflows, today announced its certification of Intel Sapphire Rapids with a benchmark of 106% improvement in GPU utilization over equivalently configured NVIDIA DGX systems connected via InfiniBand. The system was composed from 16 NVIDIA A100 […]

More than 40% of attendees believe that HPC needs composability now. Dallas, Texas, November 15, 2022 – GigaIO, the leading provider of workload-defined infrastructure without compromise for HPC + AI workflows, organized a unique panel discussion that was held today at SC22: “Smackdown! Does HPC Need Composabilty Now?” The panel was organized like a debate, with […]

Composability solutions reduce analysis time by boosting accelerator utilization while reducing costs. San Diego, California, November 3, 2022 – GigaIO, the leading provider of workload-defined infrastructure without compromise for HPC + AI workflows, today announced the expansion of a new architecture that promises to dramatically improve computational resource utilization for both genomic and CAE analysis […]

San Diego, California, October 27, 2022 – GigaIO, the leading provider of workload-defined infrastructure without compromise for HPC + AI workflows, today announced its collaboration with the San Diego Supercomputer Center (SDSC) in promoting the Prototype National Research Platform (PNRP), a $11.25M first-of-its-kind fully composable cyberinfrastructure ecosystem for the National Science Foundation (NSF). The two […]

SDSC and TACC headline community-building program at SC22. San Diego, California, October 6, 2022 GigaIO, the leading provider of workload-defined infrastructure without compromise for HPC + AI workflows, today announced the launch of its first user group, which will meet biannually at Supercomputing and ISC. This inaugural meeting marks the emergence of the composable computing […]