GigaIO to Showcase AI Innovations at ISC High Performance 2025
GigaIO, a leader in scalable edge-to-core AI platforms, will be showcasing its cutting-edge innovations at ISC High Performance 2025, taking place from June 10-13 in Hamburg, Germany.
At stand H22, attendees can experience how GigaIO’s advanced AI fabric technology bridges the gap between edge and core, providing an open and dynamic platform that supports any accelerator. This technology powers GigaIO’s flagship products, SuperNODE and Gryf, offering unprecedented scalability and performance for diverse AI workloads.
SuperNODE and Gryf: Leading the Way in AI Performance
GigaIO’s SuperNODE is hailed as the most powerful and energy-efficient scale-up AI computing platform in the world, while Gryf is a first-of-its-kind, suitcase-sized AI supercomputer designed to bring data center-class computing directly to the edge.
Both products benefit from GigaIO’s AI fabric, which allows seamless integration of GPUs and inference accelerators from leading companies such as NVIDIA, AMD, and d-Matrix. This integration enables organizations to cut power and cooling requirements by up to 30%, all without compromising performance.
AI Fabric for Faster Performance and Flexibility
GigaIO's AI fabric is based on a native PCIe Gen5 architecture, which allows for direct memory-semantic communication between distributed compute resources. This removes protocol translation overhead while maintaining sub-microsecond latencies for GPU-to-GPU data transfers.
As a result, AI workloads can achieve near-linear scaling across pooled accelerators, which are presented as though they are locally attached to the host.
Also read: OpenAI to Acquire AI-Assisted Coding Tool Windsurf for $3 Billion
Optimizing Large Language Model Training and Inference
GigaIO’s groundbreaking paper, “Rail Optimized PCIe Topologies for LLMs,” has been selected for presentation at ISC 2025 on June 12th. The paper explores optimized network architectures for large language model (LLM) training and inference, an area requiring innovative GPU interconnect solutions.
GigaIO’s rail-optimized, PCIe-based AI fabric topologies promise up to 3.7x improved collective performance and are designed to be accelerator-agnostic, ensuring they can handle a wide range of AI workloads.
Alan Benjamin, CEO of GigaIO, said, “ISC 2025 arrives at a critical juncture, as AI workloads demand unprecedented hardware resources, making optimized infrastructure essential for organizations to achieve their performance targets.”
He added, “Our expanded conference participation will demonstrate how our PCIe-based fabric technology delivers superior performance for LLM training and inference, while dramatically reducing power consumption and total cost of ownership.”
GigaIO’s expanded participation at ISC 2025 will further solidify its position as a key player in the AI infrastructure space, helping businesses optimize their AI systems for better performance and sustainability.
