The Slurm console

You can view and manage your Managed Slurm cluster by visiting the Slurm console, which you can access from the Lambda Cloud console.

Live views

The console provides live views of your cluster, including:

• Cluster health: node states, job flow, and failures at a glance.
• GPU fleet: per-GPU telemetry for every node, including temperature, power, and memory.
• Health checks: pass/fail history for every automated check on every node, with detail on what is failing and why. See Health checks for how the checks work.

Job management and observability

The console lets you work with the scheduler directly:

• Watch the live queue, filtered to your own jobs or everyone's.
• Browse job history and per-user accounting, powered by the cluster's accounting database.
• Drill into any job to see its state, its nodes, its output, and its per-GPU metrics.
• Submit new jobs straight from the browser.

User and access management

The Slurm console also lets you manage and review user access for your cluster. Available operations include:

• Adding and managing cluster users, their SSH keys, and their Slurm accounts. For details, see Manage users from the Slurm console.
• Reviewing sign-in activity and a full audit trail of changes to the cluster's control plane.