Giga Computing & Start Campus partner on AI study

Giga Computing and Start Campus have announced a joint technical study to evaluate the integration of advanced data centre technologies and infrastructure.

The study will focus on integrating Giga Computing's GIGAPOD platform and GIGABYTE POD Manager with AI-ready data centre infrastructures, such as SINES DC. The evaluation will target optimising energy efficiency, enhancing AI-driven operations, and strengthening system resilience.

GIGAPOD, noted for its ability to house up to 256 GPUs within a compact setup, will be part of the study. The configuration relies on GIGABYTE AI servers and liquid cooling technology to ensure stable operation under demanding workloads. Each GIGAPOD rack holds 64 GPUs, each consuming up to 1kW, needing only five racks for full deployment.

The system's thermal design aims to reduce energy consumption while maintaining significant performance levels. This advancement is expected to help AI-ready data centres, like those proposed, achieve elevated energy efficiency standards and lower environmental impacts.

Additionally, the GIGABYTE POD Manager software complements the hardware by streamlining operations, enhancing resource allocation, and ensuring uptime through integrated monitoring and predictive analytics. These capabilities enable data centre operators to achieve improved energy efficiency and reliable performance.

The SINES DC facility has been chosen for practical demonstration of the deployment of advanced AI workloads. Its infrastructure, capable of supporting high-density rack deployments of up to 200kW, makes it a suitable environment for solutions like GIGAPOD. The facility uses an ocean-water cooling system, which preserves water resources and achieves unmatched energy efficiency, making it a standard for sustainable high-performance AI deployments.

The study will explore deployment scenarios for Giga Computing's GIGAPOD platform at SINES DC using configurations featuring NVIDIA HGX B300 and NVIDIA HGX B200 GPUs. The study will also consider location-specific challenges and global connectivity requirements in data centre design.

Daniel Hou, General Manager of Giga Computing, stated: "Our joint technical study with Start Campus underscores our commitment to driving innovation in data centre technology. By demonstrating how GIGAPOD can seamlessly integrate into cutting-edge facilities like SINES DC, we're proving how scalable, high-density solutions will be key in enabling optimal performance and efficiency as AI workloads grow more complex."

Robert Dunn, CEO of Start Campus, added: "Engaging in this joint study with Giga Computing allows us to explore innovative data centre design solutions that meet the evolving demands of AI and cloud services. It confirms that SINES DC is ready to support the next generation of AI compute, while proving valuable insights for the design of our campus expansion. As AI workloads continue to evolve, scalable, high-density solutions like GIGAPOD will be critical in shaping the future of data centre infrastructure."

Giga Computing and Start Campus are using SINES DC's capabilities to explore potential advancements in AI-ready data centre design, showcasing their efforts and findings during NVIDIA GTC 2025.