Giga Computing Showcases Scalable AI Data Center Infrastructure at ISC 2025, Featuring Support for New NVIDIA Blackwell Ultra Platform

June 10, 2025 – Giga Computing, an industry leader in generative AI servers and advanced cooling technologies, is proud to showcase its latest advancements in AI data center infrastructure at ISC 2025, placing GIGAPOD at the heart of its exhibition. Designed for high-density AI workloads, GIGAPOD represents a leap forward in rack-scale architecture, end-to-end resource integration, and energy-efficient computing.

Unifying Power and Intelligence: GIGAPOD with GPM Integration
With the depth of expertise in hardware and system design, Giga Computing has combined AI infrastructure hardware, platform software, and architecting service to deliver scalable units composed of GIGABYTE GPU servers with NVIDIA HGX™ platforms; while running GIGABYTE POD Manager, a powerful software suite designed to enhance operational efficiency, streamline management, and optimize resource utilization. GIGAPOD’s scalable unit is designed for either air or liquid-cooled racks. Giga Computing offers two approaches for the same goal, one high-performance GPU cluster using NVIDIA Hopper and NVIDIA Blackwell GPU platforms at scale to meet demand for all AI data centers.

Direct Liquid Cooling in Action
Demonstrating leadership in thermal innovation, Giga Computing showcases GIGAPOD with its GIGABYTE G4L3 Series Direct Liquid Cooling (DLC) servers in a 4+1 rack design. Using DLC, this system boosts energy efficiency and performance at higher compute densities—key for modern AI training environments. With validated system-level integration, Giga Computing accelerates time-to-deployment through partnerships with top-tier industry partners.

AI Data Center Services from Consultation to Operation
Positioning itself as an end-to-end AI data center infrastructure builder, Giga Computing now delivers L12-level data center services that includes consulting, facility planning, system integration, deployment, and ongoing operations—backed by a robust portfolio of AI servers and rack-scale solutions compatible with NVIDIA HGX™, AMD Instinct™ and Intel^® Gaudi^® 3 platforms.

Empowering the Next Era of AI Training
GIGABYTE servers are built to meet the demands of cutting-edge AI training and inference workloads:

• NVIDIA RTX PRO™ Servers: Integrating NVIDIA MGX™-based systems with NVIDIA RTX PRO™ 6000 Blackwell Server Edition GPUs, NVIDIA Spectrum™-X Ethernet networking, and NVIDIA BlueField^®-3 DPUs - perfect for agentic AI workflows, digital twins, and enterprise AI deployments.
• AMD Platform: Featuring the latest AMD EPYC™ 9005 Series Processors, AMD Instinct MI350 Series GPUs, and AMD Pensando™ Pollara 400 NICs for exceptional AI performance.
• Intel Platform: Built with 5th Gen Intel^® Xeon^® CPUs and Intel^® Gaudi^® 3 accelerators, ideal for enterprise-scale AI deployments.

Spotlight on NVIDIA GB300 NVL72 Platform
A major highlight of GIGABYTE’s booth is the NVIDIA GB300 NVL72 platform: a groundbreaking AI inference platform featuring 72 NVIDIA Blackwell GPUs, 36 NVIDIA Grace™ CPUs, NVIDIA ConnectX^®-8 SuperNICs and full liquid cooling in a single cabinet. This solution underscores Giga Computing’s readiness to meet the scale and complexity of today’s most demanding AI models.

Giga Computing would like to invite visitors to attend its Keynote session on the topic:

“GIGAPOD: Redefining the Future Data Center by Scaling Beyond Servers and Software Stack”

• Location: HPC Solutions Forum, Hall H, Booth L01 - Ground floor
• Time & Date: 15:00 – 15:20 pm, Wednesday, 11th of June

Visit GIGABYTE at booth D40 at ISC 2025 and learn more about rack-scale data centers and how Giga Computing has grown to be a service provider for hyperscalers.