Listing Nodes

List all compute nodes in your Fuzzball cluster

select

Please select either the web UI or CLI tab to see the appropriate instructions for your environment.

web UI

As of the time if this writing, the web UI does not have the capability to display node information. This functionality is coming soon.

CLI

List all compute nodes in the cluster:

$ fuzzball node list

NODE ID          | HOSTNAME        | CPU TYPE   | TOTAL CORES | TOTAL MEMORY (GB) | TOTAL DEVICES | RUNNING JOBS
node-worker-01     worker-01.local   cpu/x86_64   8             16.0                2               1
node-worker-02     worker-02.local   cpu/x86_64   16            32.0                3               0
node-worker-03     worker-03.local   cpu/x86_64   32            64.0                4               2

The output shows:

NODE ID: Unique identifier for the node
HOSTNAME: Network hostname
CPU TYPE: CPU model name or architecture
TOTAL CORES: Total number of CPU cores
TOTAL MEMORY (GB): Total memory capacity in Gigabytes
TOTAL DEVICES: Total number of consumable devices
RUNNING JOBS: Number of jobs currently running on this node

Available Resources

Use the --available flag to show the available resources on each node:

$ fuzzball node list --available

NODE ID         | HOSTNAME         | CPU TYPE   | AVAILABLE CORES | AVAILABLE MEMORY (GB) | AVAILABLE DEVICES | RUNNING JOBS
node-worker-01    worker-01.local    cpu/x86_64   6                 12.0                    gpu/nvidia:1        1
node-worker-02    worker-02.local    cpu/x86_64   16                32.0                    gpu/nvidia:2        0
node-worker-03    worker-03.local    cpu/x86_64   28                60.0                    gpu/nvidia:4        2

JSON Output

For programmatic access or scripting, use the --json flag:

$ fuzzball node list --json

This returns node information in JSON format suitable for parsing with tools like jq:

$ fuzzball node list --json | jq '.[] | select(.running_jobs > 0)'