Joule Dashboard Generation Agent

Use Joule with AI-Assisted Dashboards to efficiently build and display graphical cards based on your SAP Cloud ALM data.

Joule AI-Assisted Dashboards are embedded in SAP Cloud ALM. You can access them in the following applications:

Monitoring Dashboards

See also the ALM Dashboard Generation Agent Information published on SAP site on SAP Discovery Center.

Data Sources

To generate cards for custom dashboards, you can use information from the following data sources:

Monitoring Dashboard app:

Real User Monitoring
Health Monitoring
Integration Monitoring
Exception Monitoring
Jobs and Automation Monitoring
Business Process Monitoring
Monitoring Events
Configuration & Security Analysis

Prerequisites

In SAP for Me, you've activated Joule with SAP Cloud ALM as described Joule with SAP Cloud ALM. You've also activated the Joule Dashboard Generation Agent.
You've created a custom dashboard. Joule with AI-Assisted Dashboards can only create cards for custom dashboards.
You need valid authorizations in SAP Cloud ALM to access its data sources. Refer to Role Collections for an overview of roles and authorizations available for project management.

Using Joule AI-Assisted Dashboards

Start the Joule AI-Assisted Dashboards as follows from the supported app, Monitoring Dashboards or Cross-Project Overview.
When you interact with the Joule assistant to create or modify charts, the card content is displayed in the context of the app:
- Monitoring Dashboards: Use the SAP Cloud ALM scope selector to choose your services or systems and select a time frame. You can select up to three systems.
Select your custom dashboard.
Open Joule in one of the following ways:
- Choose  (Joule) in the top-right menu.
- In edit mode, choose  (Joule) in the card menu. The card content loads in the Joule assistant.
Now you can start to work on AI-created dashboards.

Creating Cards in Joule

Enter a prompt indicating the data you'd like to display along with the type of chart you'd like to use. Example: Display a chart showing the CPU of my systems for the last seven days in hours.
Updating Cards

Enter a prompt to make changes to the chart. For example:
- Add a drilldown by service name.
- Display in a column chart for this month.
- Add the user response time to the chart.
Loading Cards

You can load cards in two ways:
- In Joule, enter a prompt to specify the name of the card you want to load. For example Load card My Systems CPU.
- In the SAP Cloud ALM apps, choose  (Joule) in the card menu while you are in edit mode.
Explaining Cards

Ask the Joule to explain card content by entering a prompt like Explain the cards.

Card Elements

Time Dimension

To build compelling stories with your data, analytics use a rolling time dimension with two attributes: period and granularity.

The period defines the duration of measurements, and the resolution defines the scale of data points. For example: Today/Hour or Last six Months/Month).

The time dimension propagates from dashboards to individual charts.

Two modes of period are supported:

Fixed: The period is defined at the card level and can't be inherited from the page.
Dynamic: The period is defined at the page level and inherited at the card level.

Scope Dimension

Two modes of scope selection are supported:

Fixed: The scope is defined at the card level and can't be inherited from the page.
Dynamic: The scope is defined at the page level and inherited at the card level.

Supported Charts

Charts visualize data. Some charts are dedicated to table format, while others are used for series format.

A chart can display multiple series, which need a common x-axis dimension. Usually, the x-axis is defined by the selected time period.

Charts are useful for showing relationships between multiple measures over time.

There are charts of the following types:

Trend: This chart shows trends over time. Depending on your needs, use a single or double scale.
Comparison: This chart shows comparisons between two or more categorical values.
Distribution: This chart shows the distribution of values within a data set.
Compliance: This chart shows compliance of dimensions toward an objective.
Table: The table chart shows raw data in a structured row and column format. Tables usually offer breakdown functionality with links to detailed charts.

Visualization	Type	Purpose
Time Series	Trend	Time series visualization is the default way to show the variations of a set of data values over time.
Time Series Column	Trend	Time series represented in a column.
Time Series Dual Axes	Trend	The visualization can render series as lines or bars.
Area Chart	Trend	Graphically displays quantitative data based on the line chart.
Line Chart		Display data points connected by straight lines.
Column Chart	Comparison	A column chart uses vertical rectangular bars, where the length of each bar represents each value.
Stacked Column Chart	Comparison	Columns are stacked on each other.
100% Stacked Column Chart	Comparison	Each series column represents the percentage of the overall category to which it belongs.
Bar Chart	Comparison	A bar chart uses rectangular bars, where the length of each bar represents each value.
Stacked Bar Chart	Comparison	Bars are stacked on each other.
100% Stacked Bar Chart	Comparison	Each series bar represents the percentage of the overall category to which it belongs.
Pie Chart	Distribution	A pie chart is a graph that displays data as segments of a circle proportional to the whole. Each slice corresponds to a value.
Donut Chart	Distribution	A variation of a pie chart.
Radar Chart	Compliance	Display multiple data points and the variation between them.
List Card - Tile	Compliance	Display your data in single values of interest in tiles.
List Card - Table	Compliance	Display your data in single values of interest in a table (default format).
Stat Card	Compliance	Display your data in single values of interest. Historical values are added in a form a column chart.
Status History	Compliance	Display data in a way that shows periodic states over time.
Table	Table	Visualization designed to display data in columns and rows.