Hkube Dashboard can help you in:

Accessing the Dashboard UI #


Jobs #

The Jobs tab let you monitor your pipelines state and progress.


Extra details #

Pressing the + icon near the Job ID will open the extra details panel, here you can see the DAG of your pipeline. You can also see the status, results and pipeline data.


Pipeline Details

Workers #

The Workers tab let you monitor your currently active workers.
A Worker is an Hkube process that responsible for running a single algorithm.
So this screen is basically showing details about your running algorithms.


Algorithms #

The Algorithms tab let you add, edit and delete your algorithms.


Add algorithm #

Adding algorithm is easy, just press the + button and fill the following details:

  • Algorithm name - must contain only alphanumeric, dash or dot.
  • Algorithm image - e.g. hkube/algorithm-example*.
  • CPU usage - in cores.
  • Memory usage - in Mb.

*make sure the image exists in the docker hub

Add Algorithm

Delete algorithm #

Deleting algorithm is also easy. However, if you have any stored/running pipelines that depend on these algorithms, they will be deleted as well.

Delete algorithm

Delete algorithm

Debug algorithm #

If you want to inspect your algorithm integration with Hkube, you can debug your algorithm using the debug tab.
Follow these three steps:
1) Press the + button, write the algorithm name for debugging.
2) Run a pipeline that contains this new algorithm you just created.
3) Copy the path to your WebSocket client code.

Debug algorithm

Pipelines #

The Pipelines tab let you add, edit and delete your pipelines.
You can also execute pipelines, this will let you edit some properties before execute.

Modifying the pipeline in the execute window will not affect the stored pipeline.

Stored pipelines

Execute pipeline