Unlocking Innovation for Canadian Research

Building a stable, scalable, and AI-ready future for Canadian researchers.

Industry

Education

Use Case
Artificial Intelligence

Powering Canadian Supercomputing

Overview

SHARCNET, a consortium of 19 academic institutions and a Canadian national host site, powers the Nibi cluster at the University of Waterloo. Combining over 135,000 CPU cores and 288 GPUs, it supports thousands of researchers across fields from bioinformatics to physics and the humanities.

Solution

SHARCNET chose the VAST AI OS to replace its Lustre system and serve as the data foundation for its refreshed infrastructure. The unified platform eliminates complex storage tiering to easily handle wildly different I/O patterns across diverse academic workloads.

Results

VAST quietly absorbed billions of files and demanding workloads without becoming a debugging issue. By removing single-job cluster impacts and maintenance overhead, SHARCNET provides a reliable platform for GPU-accelerated research across Canada.