<languages />


{| class="wikitable"
|-
| Availability: TBA
|-
| Login node: <!-- <b>killarney.alliancecan.ca</b> --> - TBA
|-
| Globus endpoint: TBA
|-
| System Status Page: TBA
|}

<b>Killarney</b> is a cluster dedicated to the needs of the Canadian scientific Artificial Intelligence community. <b>Killarney</b> is located at the [https://www.utoronto.ca/ University of Toronto] and is managed by the [https://vectorinstitute.ai/ Vector Institute]  and [https://www.scinethpc.ca/ SciNet]. It is named after the [https://www.ontarioparks.ca/park/killarney Killarney Ontario Provincial Park], located near Georgian Bay.

This cluster is part of the Pan-Canadian AI Compute Environment (PAICE).

==Killarney hardware specifications==

{| class="wikitable sortable"
!Performance Tier || Nodes !! Model || CPU !! Cores !! System Memory!! GPUs per node || Total GPUs 
|-
|  Standard Compute || 168 || Dell 750xa || 2 x Intel Xeon Gold 6338 || 64 || 512 GB || 4 x NVIDIA L40s 48GB || 672
|-
|  Performance Compute || 10 || Dell XE9680 || 2 x Intel Xeon Gold 6442Y || 48 || 2048 GB || 8 x NVIDIA H100 SXM 80GB || 80
|}

==Storage System==

<b>Killarney</b>'s storage system is an all-NVME VastData platform with a total usable capacity of 1.7PB.

{| class="wikitable sortable"
|-
| <b>Home space</b><br /> xxxTB total volume||
* Location of /home directories.
* Each /home directory has a small fixed [[Storage and file management#Filesystem_quotas_and_policies|quota]].
* Not allocated via [https://alliancecan.ca/en/services/advanced-research-computing/accessing-resources/rapid-access-service RAS] or [https://alliancecan.ca/en/services/advanced-research-computing/accessing-resources/resource-allocation-competition RAC]. Larger requests go to the /project space.
* Has daily backup
|-
| <b>Scratch space</b><br /> xPB total volume<br />Parallel high-performance filesystem ||
* For active or temporary (scratch) storage.
* Not allocated.
* Large fixed [[Storage and file management#Filesystem_quotas_and_policies|quota]] per user.
* Inactive data will be [[Scratch purging policy|purged]].
|-
|<b>Project space</b><br />xPB total volume<br />External persistent storage
||
* Not designed for parallel I/O workloads. Use /scratch space instead.
* Large adjustable [[Storage and file management#Filesystem_quotas_and_policies|quota]] per project.
* Has daily backup.
|}

==Network Interconnects==

Standard Compute nodes are interconnected with Infiniband HDR100 for 100Gbps throughput, while Performance Compute nodes are connected with 2 x HDR 200 for 400Gbps aggregate throughput.

==Scheduling==
The <b>Killarney</b> cluster uses the Slurm scheduler to run user workloads. The basic scheduling commands are similar to the other national systems.

==Software==
* Module-based software stack.
* Both the standard Alliance software stack as well as cluster-specific software.