Vast Data Selected to Power New Wave of GPU-Accelerated Research and Sovereign Cloud Innovation Across Canada

Vast Data, the AI Operating System company, announced accelerating momentum across Canada as leading national research centers and AI institutes standardize on the Vast AI Operating System (AI OS).SciNet at the University of Toronto, and SHARCNET at the University of Waterloo have adopted Vast as their data and AI operating system, building on previously announced deployments with Canadian-based infrastructure providers that are using Vast to power large-scale GPU services.

Together, these organizations represent a cross-section of Canada’s AI ecosystem – from national academic compute sites and world-renowned AI research institutes to cloud infrastructure providers – converging on a common requirement: a scalable, resilient data and AI operating system that can keep pace with GPU-era workloads while remaining simple enough to operate with lean teams.

SciNet, one of Canada’s five national academic HPC sites and home to the Trillium HPC, is running its third-gen system with approximately 240,000 CPU cores and 250 GPUs. Rather than repeat a legacy model, SciNet has consolidated on the Vast AI OS infrastructure, designed to serve traditional HPC codes and emerging AI workloads side by side.

“We run an extraordinary mix of workloads by researchers across Canada, and our new system pushes well beyond the scale of our previous gen,” said Daniel Gruner, CTO, SciNet. “With Vast, we no longer have to juggle different storage tiers or bolt on burst buffers just to keep up with demanding I/O. We can support both traditional HPC and new AI workloads on Vast’s AI OS that our users hammer every day – and it simply keeps up.”

SHARCNET, another of Canada’s national host sites and the organization behind the new Nibi cluster at the University of Waterloo, has similarly chosen the Vast AI OS as the data foundation for its refreshed infrastructure. Nibi combines more than 135,000 CPU cores with 288 GPUs and advanced immersion cooling, and serves a “long tail” of research disciplines – from bioinformatics and physics to economics and the humanities.

“We support thousands of researchers with wildly different I/O patterns, from billion-file bioinformatics datasets to students learning HPC for the first time,” said John Morton, director, technology, SHARCNET. “With our previous system, a single misbehaving job could impact the entire cluster. The best thing we can say about Vast is that it has largely disappeared into the background – it has quietly absorbed billions of files and some extremely demanding workloads without ever becoming a problem we have to debug.”

Vast AI OS allows Canada’s national host sites and AI clouds to: 

• Unify HPC and AI on One Platform: Vast AI OS allows organizations to run tightly coupled simulations, data analytics, AI training and inference on a single, unified platform – instead of managing multiple tiers, burst buffers, and database and compute services from disparate infrastructure
• Keep GPUs Fully Utilized: Vast’s Disaggregated Shared-Everything (DASE)  architecture delivers consistent performance for small-file, high-IO/s and metadata-heavy workloads so that CPU and GPU resources remain compute-bound, not I/O-bound
• Simplify Operations for Lean Teams: Vast provides teams with a modern management experience, built-in snapshots and catalog services, and automated resilience features that reduce the need for manual tuning and firefighting
• Enable Secure, Multi-Tenant Environments: The Vast DataSpace delivers a global namespace with multi-tenancy and data services to support secure, compliant environments for sensitive data and multi-organization research collaborations
• Prepare for Multi-Site and Multi-Cloud Futures: Vast enables organizations to build toward a common data foundation that can span national research sites, Canadian infrastructure providers and public cloud environments, enabling replication, shared datasets and consistent governance

“Canada has made bold, strategic investments in AI – from national research centers to cloud infrastructure designed for large-scale GPU computing,” said Pezhman Sharifi, director, Vast Data Canada Inc. “What unites SciNet, SHARCNET and other Canadian infrastructure partners is a shared need for an operating system that keeps GPUs and researchers fed with data without adding operational complexity. The Vast AI OS delivers a single, global platform for Canada’s innovation economy – from academic labs to production AI services.”

Learn more about the Vast’s industry-leading approach to AI infrastructure by registering for Vast Forward. Learn. Connect. Accelerate. Join VAST’s first-ever user conference, Vast Forward, February 24–26, 2026 in Salt Lake City, Utah. Hear from VAST leadership and customer and partner speakers, dive into technical breakouts, get hands-on in expert-led labs, and earn certifications to validate your expertise.