Skylar Analytics: Data Visualization and Data Exploration

The Data Visualization component of Skylar Analytics contains dashboards and charts based on data gathered by Skylar AI and SL1. To display these dashboards and charts, Data Visualization uses a ScienceLogic-hosted instance of Apache Superset. The data for the dashboards and charts includes metrics for file systems, network interfaces, and all Dynamic Applications, with more metrics planned for future Skylar and SL1updates.

The dashboards and charts in the Data Visualization component of Skylar Analytics are not compatible with SL1 dashboards, widgets, or reports.

The optional Data Exploration component enables third-party tools that use the Open Database Connectivity (ODBC) interface to access the metric data from Skylar AI. This component lets you use ODBC to connect Skylar AI data with Tableau, Microsoft BI, and other business intelligence tools.

This chapter provides a general overview of how to view the charts, graphs, and other reports in the Skylar Analytics user interface, along with tips and best practices for users of SL1 and Skylar AI.

What is Data Visualization?

Before the initial release of Skylar Analytics, SL1 stored data in a proprietary format that was not easily exported to other third-party applications for further research and insight. Skylar Analytics takes the data gathered by SL1 and Skylar AI, normalizes it, and makes it available in standard ODBC database format.

The data originates from SL1 data collectors, undergoes processing, and is then simultaneously transmitted to Skylar using the API.

Datasets are curated representations of the data in your database that let you quickly create dashboards and charts in Skylar Analytics. These dashboards and charts are based on the metrics stored in the datasets. In Skylar Analytics, each dataset contains a set of related metrics, such as server reports, which you can use to build a custom dashboard made up of one or more charts. You will not need to create new datasets for Skylar Advisor.

ScienceLogic hosts an instance of Apache Superset as an option for Data Visualization that lets you explore and view your data using business intelligence (BI) dashboards. Below is an example of one of the default dashboards in Skylar Analytics:

You can also use the Data Visualization component with your existing BI tools for your company that support ODBC; this option is called Data Exploration.

Because ScienceLogic does not own the underlying framework for the Data Visualization and Data Exploration components, ScienceLogic is not responsible for maintaining or updating documentation for third-party open-source software, including Apache Superset. For the most current and accurate information, see Additional Resources for Skylar Analytics.

Viewing Dashboards and Charts in Data Visualization

The Data Visualization component of Skylar Analytics contains dashboards and charts based on data gathered by Skylar AI and SL1.

A dashboard in Skylar Analytics is similar to a dashboard in SL1, in that they both contain a number of graphical "widgets" that display data in a variety of ways, such as pie charts, line graphs, maps, bar charts, and other visualizations. A chart in Skylar Analytics works much like a "widget" in SL1, in that a chart in Skylar Analytics is a building block that makes up a dashboard, and a dashboard can contain many charts. 

The dashboards and charts in the Data Visualization component of Skylar Analytics are not compatible with SL1 dashboards, widgets, or reports.

Unlike dashboards in SL1, a dashboard in Skylar Analytics is used only for laying out the various charts that make up that dashboard. You can use charts to customize the data. One significant difference is that a chart, when modified, impacts all dashboards using that chart definition.

As a best practice, you should make a copy of a chart if you want to modify that chart for different analyses on different dashboards.

Logging In to the Data Visualization Component

To log in to the Data Visualization component of Skylar Analytics:

1. From SL1, go to the Skylar AI page () and click the Visit button for Skylar Data Visualization. The Skylar AI login page appears.

   If you know the URL of your Skylar AI system, you can go to that location instead of using SL1.

2. Click Analytics and, if needed, type in your user name and password. The Home page for the Data Visualization component of Skylar Analytics appears:

   

   The Home page contains links to the dashboards and charts that you have used the most, including those that you have marked as favorites (). You can create a dashboard or a chart from this page, and you can view all dashboards or charts by clicking the corresponding View All link.

3. Click a dashboard or chart from the Home page, or click the Dashboards tab or the Charts tab to view a list of all dashboards or charts.

   To return to the Skylar AI login page, click the Skylar Analytics icon at top left.

4. For more information about viewing existing dashboards, see Viewing Skylar Analytics Dashboards.

5. For more information about creating or customizing dashboards, see Creating and Customizing Dashboards and Charts..

Default Skylar Analytics Dashboards

Skylar Analytics includes a set of default dashboards created by ScienceLogic that you can use to view data. You can also customize these dashboards as needed.

Each default dashboard has the word "(Sample)" or "(Skylar)" at the end of its name to show that it is a ScienceLogic dashboard, and also to remind you to duplicate any of these dashboards or charts if you wish to make modifications. They are also owned by the System Administrator ("SA") user. These SA-owned dashboards and charts might be updated by ScienceLogic periodically.

The Dashboards tab for Skylar Analytics contains the following default dashboards:

• Filesystem Overview + Exploration (Sample).

  • Displays 95th percentile data, file system utilization distribution (as a percentage and Gigibit or GiB), and "Space Hogs" (the devices using the most file system space). 
  • You can click a device name on the "Space Hogs" pie chart to display chart details specifically for that device.
  • Also includes the Ad-Hoc Comparative Analysis tab, which displays additional file system charts for all devices or selected devices from the Overview tab.

• Filesystem Statistics (Sample). Displays a pie chart of "Space Hogs" (the devices using the most file system space), file system utilization as a percentage, file system inventory by host, and file system usage distribution.

• Filesystem Usage (Sample).

  

  • Displays a set of file system usage, utilization, 95th percentile and Top-N inventory charts for all devices, including a pie chart of "Space Hogs" (the devices using the most file system space).
  • You can click a device name on the "Space Hogs" pie chart to display chart details specifically for that device.

• Interface Statistics (Sample). Displays interface traffic in a variety of charts, including active hosts, active interfaces, dropped packets, and 95th percentile for the last 30 days (as a percentage and MIBPs).

• Most Significant Resource Changes (Sample).

  • Displays devices with the highest delta of file system usage, along with average file system usage, Top-N interface usage delta, and interface traffic in the past seven days.
  • You can click a device name on the "Top-N Filesystem Usage" or the "Top-N Interface Usage" tables to display chart details specifically for that device.  

Viewing Skylar Analytics Dashboards

You can use the following tips to get more data from your Skylar Analytics dashboards:

• For most dashboards, you can click a single device or item in the first chart at the top left of the Dashboard page (or any "Top-N" chart types) to view data specific to just that device. Click the device a second time to clear the filter.
• Hover over a graphical element in a chart, such as a piece of a pie chart or a colored metric in a tree map to view a pop-up with more information about that element.
• If a dashboard is editable, you can click Edit Dashboard to make changes to the dashboard and the charts that make up the dashboard. For more information, see Creating and Customizing Dashboards and Charts.

The following image displays a dashboard with a device selected in the "Space Hogs" graph that forces the other graphs to only display data for that device:

When viewing a dashboard, you can click the ellipsis button () at the top right of the Dashboard page to open a menu with the following dashboard options:

• Refresh dashboard. Updates all of the charts in the dashboard to account for any changes you might have made.
• Enter fullscreen. Displays the browser window containing the dashboard display as full screen. Select Exit fullscreen from the menu to return to the previous setting.
• Save as. If a dashboard is editable, lets you save a copy of the dashboard, with the option of overwriting the existing dashboard or changing the name to make a new dashboard (if you have appropriate permissions).
• Download. Lets you export the dashboard as a PDF or download the dashboard as an image.
• Share. Lets you copy a link to the chart to the clipboard of your computer, and also lets you share a link to a chart using email.
• Set auto-refresh interval. Lets you choose how often you want Skylar Analytics to update the data for the dashboard. The default is Don't refresh.

On a Dashboard page, you can also click the vertical ellipsis button () at the top right of a chart on the dashboard to open a menu with the following chart options:

• Enter fullscreen. Displays the browser window containing just this chart display as full screen. Click the Exit fullscreen icon () or select Exit fullscreen from the menu to return to the previous setting.
• Edit chart. Opens the Edit Chart page so you can add metrics, edit queries, and make other updates to this chart. Click Save to keep your changes (if you have appropriate permissions).
• Cross-filtering scoping. Lets you add cross-filtering, which lets you apply a data element from a chart (like a table row or a slice from a pie chart) and then apply it as a filter across all eligible charts in the dashboard.
• View query. Displays the SQL query for that chart. You can use this option to determine which data from the dataset is being used in this chart, and how the data is being used.
• View as table. Displays the chart in table format.
• Drill to detail. Displays all the data that makes up a chart.
• Share. Lets you copy a shareable chart link to your system's clipboard, or launches your system's default email client and composes a new message featuring the chart URL.
• Download. Lets you export the chart to .CSV or Excel, or you can download the chart as an image.

Creating and Customizing Dashboards and Charts

You can create a new dashboard in Skylar Analytics, or you can customize any of the default dashboards and save them with a new name. You can also create and customize the charts that make up the various dashboards.

To create a dashboard:

1. Log in to the Skylar AI user interface and click Analytics. The Home page for Skylar Analytics Data Visualization appears.

2. In the Dashboards section of the Home page, click the +Dashboard button. An Untitled Dashboard page appears.

3. Triple-click in the [untitled dashboard] field at the top right and type a name for the new dashboard. If you are using a shared system, you might want to add your initials to the end of the name.

4. Click Save in the upper right corner of the page.

5. Click Edit the Dashboard.

6. If there are existing charts that you want to add to this dashboard, click and drag each chart from the Charts tab on the right and drag the chart onto the dashboard. Click Save when you are done, and click Edit Dashboard again to keep editing.

   If you want to see only the charts that you have created, check Show only my charts. If you want to see charts by all users, clear this option.

7. If you have not yet created any charts, or no charts existing on your system from other users, click Create New Chart. The Create a new chart window appears:

   

8. In the Choose a Dataset field, click to choose a dataset with the data you want to view in your new dashboard. In Skylar Analytics, a dataset contains a set of related metrics pulled from Dynamic Applications in SL1, such as server reports or SL1 business service statistics. 

   For this overview, we will select the BusinessServiceStatistics dataset, which contains data about SL1 business services.

9. Select a chart type from the Choose chart type section and click Create New Chart. For this overview, we will select Area Chart. A new chart window appears:

   

   In the first column, the Chart Source field displays the dataset you selected (for this overview, it is the BusinessServiceStatistics chart source). Below that field, you can access the metrics and columns that you can add to the chart. You can drag a metric or column from the first column into the second column to add it to the chart.

   In the second column, you can select which data will appear in the chart, and how the data will be displayed in the chart. The large section to the right displays a preview of the chart as you build it after you click the Create Chart button to run the query.

10. For example, with an Area Chart type, you could define the X-axis of the chart to display a time range by clicking in the X-axis field in the Query section. A modal appears:

    

11. On the Simple tab of the modal, click the Column field to get a list of data. You can pre-filter the data by typing a column name or label, such as "time".

12. Select a column with a calendar icon () next to it to display a time range on the X-axis, such as BsvcDataTime from the BusinessServiceStatistic chart source, and click Save.

    For more information about the abbreviations used for the metric names, see Mapping SL1 Dynamic Application Object Names to Skylar Columns.

13. To set the granularity of the time frame to a shorter time frame, click in the Time Grain field and select Hour instead of the default of Day.

14. Next, select the metrics you want to visualize in the chart by clicking in the Metrics field and clicking the Simple tab. You can also drag a metric from the first column and drop it on this field.

    For this overview, we will select BsvcHealth (business service health) in the Column field and AVG in the Aggregate field. We will also select BsvcRisk (business service risk) in the Column field and AVG in the Aggregate field. These settings will show the average values for business service health and risk over time.

    In the Column field on the Simple tab, type % to filter the list down to Utilization or Percentages.

15. Click Save to save the metrics.

16. To see a preview of the chart so far, click the Update Chart button or the click here link in the large section to the right. You will need to do this every time you make a change if you want to see the latest preview.

17. You can use the Dimensions field to add descriptive elements to the chart that help users understand the data being visualized. For example, for line charts and area charts, the dimensions will appear in the legends and mouseover text. For tables, dimensions represent the columns to display.

    For this overview, we will add BsvcName (business service name) to the Dimensions field. Click the Update Chart button to see an updated preview:

    

    The chart legend displays the average health and average risk in the legend at the top, and also in the columns at the bottom of the section. If you mouse over a line in the chart, you can see the specific data for those values.

18. In the Filters field, you can edit the existing filter by clicking on it and specifying what data to display on the new chart. The filter currently has no specific filter set right now.

    For this overview, click on BsvcDataTime in the Filters field and then click in the Time Range field. An Edit time range modal appears.

19. In the Range Type field, select a range, such as Last, as in Last day, Last week, and so on.

20. Select the Last week option and click Apply.

21. Click the Update Chart button to review your updates.

22. To finish the chart, be sure to give it a name in the top right of the window, and then click the Save button.

23. In the Save chart modal, click Save & Go to Dashboard. The new chart is added to your dashboard.

24. Continue adding charts to the dashboard as needed.

25. When your dashboard is complete, click the Draft button at top left to publish it. The button changes from Draft to Published.

    For more information, see Creating Your First Dashboard and registering a new table in the Superset documentation.

To customize a dashboard:

1. Select the dashboard from the Dashboards page. You can also hover over the dashboard and click the Edit icon in the Actions column.

2. On the Dashboard page, click Edit Dashboard. The Charts and Layout Elements tabs appear.

3. If there are existing charts that you want to add to this dashboard, click and drag each chart from the Charts tab on the right and drag the chart onto the dashboard. Click Save when you are done, and click Edit Dashboard again to keep editing.

   If you want to see only the charts that you have created, check Show only my charts. If you want to see charts by all users, clear this option.

4. If you want to add extra elements to your dashboard, like a header, additional text, a divider, or other items, drag and drop the items onto the dashboard from the Layout Elements tab. Click Save when you are done, and click Edit Dashboard again to keep editing.

5. To edit a chart in the dashboard, click the vertical ellipsis button () at the top right of the chart on the dashboard and select Edit chart. For more information, see steps 8-21 in the "To create a dashboard" procedure.

6. When you are done updating the dashboard, you might need to click Edit Dashboard and rename the dashboard if the dashboard was created by ScienceLogic, with the word "(Sample)" or "(Skylar)" at the end of the name.

On the Dashboards tab in Skylar Analytics, the "Visualization Variety Testing (Sample)" dashboard contains a variety of chart visualizations related to file system utilization, including a table, a "big number" with a line graph, a gauge, a set of tree maps, and a sunburst map. You can use this dashboard to see how these different types of charts might work for your data.

Icons for Chart Metrics

Each data type includes a small icon that conveys its type:

• ƒ: Function used for metrics
• : The time column for the data source
• abc: Text data
• #: Numeric value data

Data Exploration: Exporting Data from Skylar AI

You can use the optional Data Exploration component of Skylar Analytics to enable Open Database Connectivity (ODBC) to connect Skylar AI data with third-party tools like Grafana, Power BI, Tableau, Cognos, Crystal Reports, SAP, Excel, and other business intelligence applications.

You can also import data from third-party tools, such as billing data, environmental data, or service level objectives (SLOs), and then use that data in Skylar AI.

Data Exploration lets you view Skylar AI data alongside other business sources, offering a holistic perspective on your operations.

Configuring Data Exploration with Power BI

This section covers how to set up an ODBC connection for Skylar Analytics so you can use it with Power BI for data visualization. Other business intelligence applications will use a similar process to integrate with Skylar Analytics.

To install and configure the ODBC connection:

1. Go to the ClickHouse ODBC driver releases page at https://github.com/ClickHouse/clickhouse-odbc/releases.

2. Download the relevant version for your operating system.

3. Open the ODBC Data Source Administrator application.

4. On the User DSN tab, click Add. The Create New Data Source dialog appears:

   

5. Select ClickHouse ODBC (Unicode) and click Finish. The Create data source for Clickhouse dialog appears:

   

6. Complete the following fields with ODBC connection details from the ScienceLogic Site Reliability Engineering (SRE) team:

• Name: Add a name to identify this connection. This will be used later in the BI tools.
• Host: Specify the host URL, provided by SRE.
• Port: 443.
• Database: Leave blank.
• SSLMode: Type the word "require".
• User: dataviz
• Password: Specify the password, provided by SRE.

To connect your BI tool, such as the Power BI Desktop:

1. Launch the Power BI Desktop and click Blank Report.

2. Click Get data from another source, select Other, and then select ODBC.

3. Click Connect.

4. In the pop-up window, click the drop-down menu and select the ODBC connection you just created in the previous procedure.

5. Click OK.

6. If prompted, re-enter your username and password, and then click Connect.

7. After you are connected, a menu will appear displaying available datasets, which you can use to create dashboards in your BI tool:

   

To import data and create a dashboard with Skylar AI data in Power BI:

When selecting datasets to import, choose only the necessary tables to optimize performance. The following procedure creates a sample dashboard in Power BI.

1. On the Home screen of the Power BI Desktop, click New Visual.

2. Select a Line Chart as an example.

3. To configure the X-Axis, expand the fsstattab dataset from the right-hand Data Column.

4. Drag fsstat_ts (Timestamp) to the X-Axis in the Visualizations panel.

5. Remove the options for Year, Quarter, and Month, keeping only Day:

   

6. To configure the Y-Axis, drag fsstat_used_pct_psec (Used Percentage Per Second) to the Y-Axis.

7. To customize the data fields, click the drop-down arrow next to the selected data field. You can rename the field or modify how the value is calculated:

   

8. Continue adding additional charts and visuals as needed to finish up your dashboard.

Additional Resources for Skylar Analytics (Apache Superset Training)

This section has been provided as an independent study guide to help you identify and develop basic knowledge and skills to build data visualizations within Skylar Analytics user interface.

The following videos from ScienceLogic cover some of the key features of Data Visualization and Exploration:

• How to create your first Dashboard
• How to add cross-filtering to a Dashboard
• Datasets Overview
• Tips & Tricks: Friendly Column Names
• BYO Tool and Data: Connect via ODBC and Import Data
• How to connect PowerBI to Skylar AI via ODBC
• BYO Data - writable schema, Joins and Views!

Apache Superset-related documentation: https://superset.apache.org/docs/intro

ScienceLogic recommends the following resources for a deeper understanding of Apache Superset:

• https://www.udemy.com/course/apache-superset-for-data-engineers-hands-on/
• https://www.youtube.com/watch?v=znnmco3eK-M&list=PLzRV_ObjEwmNhRjhMNcvcDP7ZDjOXtodd
• https://superset.apache.org/community

Because ScienceLogic does not own the underlying framework for the Data Visualization and Data Exploration components, ScienceLogic is not responsible for maintaining or updating documentation for third-party open-source software, including Apache Superset.