Home
News
Tech Grid
Interviews
Anecdotes
Think Stack
Press Releases
Articles
  • Enterprise AI

Sharon AI to Deploy 1K NVIDIA B200 Cluster in Australia


Sharon AI to Deploy 1K NVIDIA B200 Cluster in Australia
  • by: Source Logo
  • |
  • February 2, 2026

SharonAI Holdings Inc., a leading Australian Neocloud, announced the planned deployment of a 1,000-unit NVIDIA B200 GPU cluster at NEXTDC’s Tier IV M3 Data Center in Melbourne, Australia.

Quick Intel

  • Sharon AI plans to deploy a 1K NVIDIA B200 cluster at NEXTDC M3 in Melbourne, marking a major advancement in sovereign AI infrastructure in Australia.
  • The deployment leverages Lenovo TruScale infrastructure and VAST Data’s high-performance storage fabric for scalable, high-density AI workloads.
  • NEXTDC M3 offers purpose-built power density, advanced cooling, and resilience optimized for large-scale Blackwell GPU systems.
  • B200-based DGX Blackwell platform delivers up to 3X training and 15X inference performance over previous generations for LLMs and complex AI tasks.
  • Integration eliminates traditional AI lifecycle bottlenecks, enabling seamless develop-to-deploy pipelines for enterprises, government, and research.
  • Sharon AI expands its GPU fleet with B200, with future plans for B300 and GB300 systems to support Asia-Pacific hyperscale and sovereign AI needs.

Deployment Details and Infrastructure Partners

SharonAI Holdings Inc. and its subsidiaries (“Sharon AI”), a prominent Australian Neocloud (SHAZ:OTC Markets, SHAZW:OTC Markets), revealed the expected rollout of a 1,000-unit NVIDIA B200 GPU cluster at NEXTDC’s Tier IV-certified M3 Data Center in Melbourne. This initiative represents a key milestone in establishing sovereign, large-scale AI compute capacity within Australia, addressing growing demand for secure, regionally controlled accelerated computing resources.

NEXTDC’s Role and Facility Capabilities

“NEXTDC is pleased to be expanding our engagement with Sharon AI, Lenovo and VAST Data as we deliver the latest generation of accelerated computing infrastructure in Australia,” said David Dzienciol, Chief Commercial and Customer Officer at NEXTDC. “The NEXTDC M3 facility is purpose-built to support this 1K B200 cluster, with the power density, advanced cooling and operational resilience required for large-scale AI workloads.”

The M3 data center’s design supports the extreme requirements of next-generation AI systems, ensuring reliable performance for data-intensive training and inference operations.

Performance Advantages of Blackwell Platform

As enterprises shift toward more complex, data-heavy AI development, the DGX Blackwell platform provides a substantial leap forward. Each system features eight Blackwell GPUs connected via fifth-generation NVLink, delivering up to 3X the training performance and 15X the inference performance compared to prior generations. This architecture excels in massive-scale applications such as large language models (LLMs), advanced recommender systems, and real-time conversational AI.

Integrated Solution with Lenovo and VAST Data

The deployment operates under Sharon AI’s existing Lenovo TruScale agreement, enabling flexible and rapid scaling of the 1,000-unit B200 cluster. Combining Lenovo’s high-density server technology with VAST Data’s high-performance storage fabric creates a unified pipeline that removes common bottlenecks in AI development and deployment cycles.

“This engagement with Sharon AI and NEXTDC marks a significant step in delivering sovereign AI infrastructure at scale in Australia. The 1,000-unit B200 cluster at NEXTDC’s M3 facility provides the foundation Australian organisations need to compete globally," said Damien Fox, country manager Australia and New Zealand, VAST Data. "VAST Data’s storage fabric and InsightEngine remove the silos and latency that typically limit large AI systems, allowing NVIDIA Blackwell to operate as intended. The result is sovereign compute that enterprise, government, and research organisations can rely on to build and run serious AI workloads across the region.”

Strategic Expansion and Future Plans

This addition significantly strengthens Sharon AI’s existing GPU portfolio, which already encompasses NVIDIA A40, L40s, H100, and H200 architectures. Looking ahead into 2026, the company plans to introduce B300 and GB300 systems, reinforcing its role as a leading provider of cutting-edge compute for hyperscale, enterprise, government, and research clients across the Asia-Pacific region.

"This expansion is about more than just raw capacity; it is about providing the specialized, sovereign architectural depth that the next era of AI requires," said James Manning, CEO and Chairman of Sharon AI. "By bringing 1,000 B200 GPUs to Australia, we are ensuring that AI natives, research, enterprise and government organizations in the Asia-Pacific region have immediate access to world-class compute."

About SHARON AI

SharonAI Holdings Inc. (“Sharon AI”) and its subsidiaries, a leading Australian Neocloud, is a High-Performance Computing company focused on Artificial Intelligence and Cloud GPU Compute Infrastructure. Our cloud GPU platform and compute infrastructure is accelerating the build of AI factories and sovereign AI solutions, powering the next wave of accelerated computing adoption.

  • Sovereign AIAI InfrastructureCloud Computing
News Disclaimer
  • Share