Navigating the dashboard
The QruiseOS dashboard makes it easy to keep track of all your QPU data. From high-level overviews of devices and workflows to the results of individual experiments, it provides a clear, easy-to-use interface for exploring and understanding your quantum systems.
Let's explore it a little.
Home¶
The dashboard home provides a quick overview of available QPUs on the left-hand side and the most recently run workflows on the right-hand side.
System Info¶
Selecting a QPU takes you to its System Info page, which displays two panels:
- the QPU description & specs panel contains relevant information about the QPU. This can be edited by admins in Admin Settings → QPUs.
- the Layout panel shows the chip layout. Each qubit is colour-coded according to the latest value of the selected characterisation metric, such as \(T_1\), frequency, and anharmonicity. You can hover over each qubit to see the current value of that metric.
Experiment Database
You can click on Experiment Database to delve more deeply into the experiments that yielded these qubit parameters.
Workflows¶
The Workflows pane lists all currently defined workflows for the selected QPU.
Workflow runs¶
When selecting a specific workflow, you see all the runs of this workflow with their corresponding start and end times, their status, and the number of completed tasks. The status is defined according to the following:
● Completed: the workflow finished running with all tasks completed successfully.
● Completed with errors: the workflow finished running but not all tasks were successful, e.g. due to a failed analysis.
● Failed: the workflow did not complete due to a software or system error.
● Cancelled: the workflow was cancelled by the user (via the Cancel button on the dashboard or with
ctrl-C in the terminal).
You can see these five statuses in the screen capture below.
When you select a workflow run, there are two ways to view it: Sequential or Graph.
In the screen capture above, you can see the Graph view, which displays the workflow in the form of a directed acyclic graph (DAG), with each individual task shown as a separate node. The blue shaded areas represent subflows, which can be collpased or expanded as desired. (See Further workflow customisation to learn more about subflows). The chronological order runs primarily from top to bottom.
In Sequential view, the workflow is displayed as a nested list of subflows (e.g. Subflow: experiment) and tasks (e.g. Task/readout-discriminator-train+Q3-7). The chronological order runs primarily from left to right.
In both views, the subflows and tasks are colour-coded to show their status: green for successful, red for failed, and blue for pending. This makes it easy to identify which tasks have completed, which have failed (e.g. interrupting the workflow), and which are still queued. The view updates automatically as the workflow progresses, allowing you to monitor task completion in real time.
You can click on any node to view the task status, start and end times, relevant plots, and the source notebook. You can directly open the notebook to see exactly how the task was executed, which is useful for debugging and checking configuration.
Experiment Database¶
For each QPU, you can click on the Experiment Database to view all past experiments performed on that QPU and sort them by experiment name or start time. You can also filter by experiment type or use the search bar to look up experiments by these fields, as well as by qubit or experiment ID. This makes it easy to track your QPU’s behaviour over time.
You can select two experiments and compare them side-by-side by clicking the "Compare experiments" button. This opens a new page displaying the metadata (name, type, creation date, description, duration, branch, targets) and plots for the two selected experiments.
Tip
If you want to load raw and analysis data from a particular experiment, you can just copy/paste its ID and use session.load_document(ID).
In the case of the \(T_1\) experiment above, the complete snippet would be: