Webinar 2025 Migrating to the upgraded national systems

Most of the Alliance national systems have been undergoing a major upgrade during this spring and summer. They have been effectively rebuilt from scratch using up-to-date hardware, bringing a significant increase in computing capacity, memory, and storage to the users. The upgraded clusters have new names: Graham became Nibi, Beluga - Rorqual, Cedar - Fir, Niagara - Trillium. (The remaining cluster - Narval - is not having an upgrade this cycle.) Some of the upgraded systems are already online, not at a full capacity yet. The plan is to make all of them available by the end of July.

This webinar is to address the concerns and questions the users of the existing systems might have. How will the upgrade affect my workflow? Are there significant changes to the way the job scheduling and file systems operate in the new clusters? How to make the best use of the increased computing capacity of the upgraded clusters, in particular in regards of the computing nodes becoming much "fatter" (many more cpu cores per node)? As the most significant change will happen to the way the GPU computing is done, the webinar will cover this in detail.

There will be time for questions at the end, please bring your questions and concerns.