The Evident ClearStone Live Real-Time Dashboard  is primary user interface where users will access monitor and manage the remotely managed systems. This is a versatile tool that enables users to customize visualizations for any metric collected by the system. Users can create multiple perspectives that may consist of performance metrics from one or more disparate data sources. If the appropriate management plugins are deployed, administrative users can control, configure, and manage remote the systems.

Accessing the Real-Time Dashboard

The Real-Time Dashboard is accessible from the main web page of the ECSL product. The application requires a web browser with an Adobe Flash Player plugin which can be downloaded from Adobe's website.

1. Locate the “Real-Time Dashboard” link in the “Software” menu on the Evident ClearStone Live home page.
   
2. Or accessed directly via application URL: http://localhost:8080/rtdashboard
3. The default administrator username is "admin" with the default password as "mattison".

Note: Replace “localhost” in the URL with the fully qualified hostname of the ECSL server where "Data Services" option was installed.

About the Real-Time Dashboard

The Real-Time Dashboard user interface can support multiple customizable visualizations that can be grouped together into "perspectives". Each perspective is represented as a tab. Within a single perspective, users can choose from a set of pre-defined layouts that contain one to four visualizations. Each visualization is configured for a single data type and chart. A single perspective can support multiple visualizations across multiple heterogeneous data sets. The following illustration highlights the layout and features of the Real-Time Dashboard.

The Real-Time Dashboard's toolbar consist of the following menu options:

• Perspective
  • New - create a new perspective
  • Open - load a shared perspective
  • Remove - remove the current perspective from the dashboard
  • Save - save the current perspective
• Operations
  • Show Results - displays the results panel for any executed operation
  • Management operations exposed by deployed Management Plugin Servers

Throughout this document, users will learn how to use these basic features to navigate and build custom dashboards. The following tutorial consists of various examples addressing key features of the Real-Time Dashboard. This tutorial assumes ECSL is configured to gathering JMX data from one or more JVMs. Some examples may refer to data types such as Java Platform Mbeans or Oracle Coherence Nodes.

Note: Not all deployments of ECSL will include the mentioned data types. Domain specific data types will vary based on licensed features and deployed management packs for different operating environments. Additional management operations depends on the deployment and configuration of the Management Plugin Servers for the remotely managed environment.

The following exercises demonstrates how to create new and manage perspectives.

Create a new perspective

1. To create a new perspective, go to the Perspective menu and select New.
2. Select from one of the following types of perspectives:
• Single Data Type - all visualizations in the new perspective will be locked into a specific data type. This is useful for analyzing multiple metrics for the same data set. All the visualizations will be filtered, grouped, and displayed using consisting data.
• Multiple Data Types - each visualization in the new perspective will be fully customizable. This is useful for view metrics from disparate data sets. The data for each visualization will be configured independently.
• Dual-Axis History Chart - a single flexible visualization used for creating overlays of disjointed data sets over a common time period. This perspective is not live.
• Tree Map - a generic tree map visualization used for analyzing two dimension of and data displayed in a hierarchical (tree-structured) data as a set of nested colored rectangles. When the color and size dimensions are correlated in some way with the tree structure, one can often easily see patterns that would be difficult to spot in other ways.
3. Upon selection, the Real-Time Dashboard creates a new tab with a default label called "New Perspective".
4. Double-click on the label to rename the perspective to "JVM".
5. At this time, click the "Save" button to save the empty perspective.
6. Once a perspective is saved, it is available to other users of the Real-Time Dashboard.

Note: A perspective that has an asterisk (*) next to the title indicates the perspective has changed from the original version. To save these changes, click on the "Save" button.

Close a new perspective

1. A perspective can be removed from the Real-Time Dashboard by selecting on the tab and clicking on the "x" icon in the tab.
2. This does not permanently remove the perspective from ECSL.

Open an existing perspective

1. To open an existing perspective, go to the Perspective menu and select Open.
2. Select from a list of available perspectives (i.e. JVM) and click "Ok".
3. This loads the selected perspective into a new tab.
4. Some perspectives have a lock symbol which indicates, they are "read-only". These perspectives cannot be updated. However, users can save a read-only these perspective with a different name.

Configuring Views

The following exercises demonstrates how to configure view(s) within a perspective. A perspective can consists of up to four views; each can be configured with different charts and disparate data.

Configuring a Bar Chart view

1. By default, newly created perspectives displays a single view.
2. Select the empty view to make it the current view (highlighted with a yellow border).
3. In the left control panel, select the Data Type drop-down list. In this example, select "JVMPlatform" from the list.
4. This automatically loads the current unfiltered JVMPlatform data records into the data panel at the bottom of the perspective.
5. Select the Visualization Type as "Bar Chart". The current view is updated with an empty bar chart.
6. Choose a JVMPlatform metric to display (i.e. Heap Memory Used). The empty view is now updated with bars that represent the "Heap Memory Used" of each JVM.
7. Mouse-over any bar to obtain more information about each bar in the appearing tooltip.
8. Use the vertical scroll bar to scroll the chart.
9. Click the "Save" button to preserve this perspective.

Adding Line Chart view

1. By default, newly created perspectives display a single view.
2. To change the default layout with additional views, click on the "Customize" button. Choose the layout that contains two columns.
3. The layout of the perspective changes and a second empty view appears. Select the empty view (highlighted with a yellow border).
4. If this was a Multi-Data Type perspective, change the Data Type to "JVMPlatform".
5. This automatically loads the current unfiltered JVMPlatform data records into the data panel at the bottom of the perspective.
6. Select the Visualization Type as "Line Chart". The current view is updated with an empty line chart.
7. In this view, change the Metric to "CPU Utilization". The view is updated with multiple line series of CPU utilization trends for the top 5 JVMs.
8. Users can increase the # of series by expanding the Data Window sliders located at the top of the chart.
9. Users can scroll back in time using the horizontal scroll bar below the x-axis.
10. The legend of the line chart appears at the bottom of the x-axis.
11. Click the "Save" button to preserve this perspective.

Grouping and Filtering

The following examples demonstrates how to organize the real-time data within a perspective. If the perspective is a Single Data Type perspective, then all the views share the same data set, therefore the grouping and filter will apply to all the views. Otherwise for a Multi-Data Type perspective, each view's data grouping and filtering can be configured independently.

The grouping feature is useful for data sets that have multiple "key" attributes that can be used for grouping related data. Another functionality to reduce data volume is filtering. Users can set value filters on multiple attributes within the data records. Using the combination of grouping and filtering, users can organize and limit the scope of the data in the dashboards.

Grouping data

1. Open or create a perspective.
2. Select a view where there are data records.
3. The data panel at the bottom of the perspective contains a data selection tree panel and a table that displays the records.
4. The data selection tree panel lists all the entries organized by the default grouping configuration as defined in the information model for this data type.
5. To group the records in a different manner, click on the "Grouping" setting in the left control panel.
6. A list of grouping attributes is presented in the first list. The second list is the current grouping method.
7. Attributes can be dragged from one list to the other. Alternatively, selecting the attribute and clicking on the left or right arrow buttons moves it from one list to the other will do the same.
8. For each entry that appears in the second list, the data selection tree panel will group the records based on the selected grouping attributes.
9. Click Apply to activate the grouping. The selection tree panel is reorganized based on the ordering of the selected grouping attributes.
10. Here is an example of a grouping attributes for Oracle Coherence Nodes.

Filtering data

Data in any view can be filtered based on specific attributes in the data set. Users can apply one or more filters on any of attributes of the data. A filter is defined using any of the following operations:

• contains - partial match of a string
• ends with - match the end of a string
• equal - exact match
• regex - regular expression
• starts with - match the beginning of a string

1. Open or create a perspective.
2. Select a view where there are data records.
3. The data panel at the bottom of the perspective contains a data selection tree panel and a table that displays the records.
4. To filter records, click on the "Filtering" setting in the left control panel.
5. The filtering panel contains an empty table of filtering rules.
6. Click the Add ("+") button to create a new filtering rule.

• In the first column select the attribute to filter by.
• In the second column, choose the matcher for the filter rule (contains, ends with, equal, regex, starts with).
• In the third column, enter the pattern to use for the filter.

7. Click Apply to activate the filtering. The data records for the current view is reduced to a smaller set based on the records that satisfy the filter criteria.
8. Additional filtering rules can be added if desired.
9. Here is an example of a filtering for Oracle Coherence Nodes.

Data Panel

The data panel for each view displays the data records for each individual view. The data panel consists of a data selection tree panel that contains an organized list of record identifiers based on the grouping configuration. The right tabular panel contains the records that display the values of each record's attributes and metrics.  
There are two techniques for filtering the records within the tabular panel:

1. Creating filter rules using the Filtering settings as described above.
2. Selecting / deselecting entries in the data selection tree panel.

The columns within the tabular panel can be rearranged by dragging the column by the header and moving it to either the left or right of other columns. Clicking on a column will also sort the rows by that column. Additional attributes or metrics can be added or removed from the table by opening the Column selection panel. In this panel, the attributes/metrics that are checked will be displayed in the table.

Tree Map Perspective

The Tree Map perspective displays hierarchical (tree-structured) data as a set of nested rectangles. In ECSL, these rectangles can represent any metric data collected or computed by ECSL. Each branch of the tree is given a rectangle, which is then tiled with smaller rectangles representing sub-branches. A leaf node's rectangle has an area proportional to a specified dimension (i.e. JVM allocated max heap) on the data. This area dimension is configurable based on the record type selected by the user. The leaf nodes can be colored according to user specified dimensions of the data (i.e. JVM used heap). This color dimension is also configurable based on the record type selected by the user. When the color and size dimensions are correlated in some way with the tree structure, one can often easily see patterns that would be difficult to spot in other ways.

To create a tree map perspective, go to the Perspective menu and select New -> Tree Map. The blank perspective will appear as follows:

To configuring the tree map, click on the Tree Map toggle button. The following dialog is presented.

Perform the following steps in this Tree Map configuration dialog:

1. Select the appropriate data type to apply to this tree map.
2. There are two levels of grouping for the tree map. The first one is an outer grouping and the second one is the inner grouping. Not all attributes of a data type are groupable, only those that have be configured as groupable in the information model are available for selection.
3. For each grouping, a filter can be specified to limit the entities to be displayed in the tree map.
4. The next dimension is the Cell Size ("area size"). This proportion of the cell sizes is baaed solely on the values of the selected metric. The values for the metrics can be filtered to only include the records with values that satisfy the metric filter.
5. The last dimension is the Cell Color. This proportion of the cell sizes is baaed solely on the values of the selected metric. The values for the metrics can be filtered to only include the records with values that satisfy the metric filter.
6. For the cell color configuration, a 3-tiered color gradient can be applied to absolute values or percentage values.
7. Lastly, additional filtering on other attributes can be configured for finer resolution of the data.
8. Once the configuration is ready, click Apply to update the perspective. Clicking Ok will apply and close the dialog.

As with any perspective, the tree map perspectives can be named and saved for sharing and future use.

Dual Axis History Chart Perspective

The Dual Axis History Chart perspective provides additional flexibility in creating composite charts by overlaying disjointed data sets over a common time period. This feature is useful for analyzing performance data from disparate data sets of one or more resources. The perspective loads data that is specified in the Time Range and Offset selection. Once data is loaded, the chart does not update automatically with any live data.

To create a Dual Axis History Chart, go to the Perspective menu and select New -> Dual Axis History Chart. The blank perspective will appear as follows:

There are two Chart Axis buttons. One for configuring the primary Y axis and series and the second one for configuring the secondary Y axis and series. Proceed to configure the primary Y axis by clicking on the first Chart Axis button. The following dialog will be displayed:

Perform the following steps in this Primary Y Axis Options dialog:

1. Select the appropriate data type to apply to this axis.
2. Choose one of the Axis Type:
• Single Metrics with Multiple Instances - allows only one metric to be selected for multiple unique entities. Each series represents a single unique entity.
• Multiple Metrics for Single Instance - allows multiple metrics to be selected for a single entity. Each series represents different metrics for the same entity. The metrics all must have the same unit of measurement.
3. Choose the line style for the line chart.
4. Select the one or more metrics (based on step 2).
5. For the instance selection, a text filter can be supplied to limit the unique instances that match the filter.
6. Select one or more instances (based on step 2).
7. Clicking Ok will apply and close the dialog.
8. At this point, the chart is ready to be populated with data. The user can specify the time range and offset of data to load. The availability of the data is limited to how much data is actually store in the ECSL OpCache.
   
   
   • The range is from 5 mins to 6 hours of data.
   • The offset is relative to the current time by 30 minute increments.
9. Once these options have been selected, click on the Load/Refresh button to retrieve this data.

Optionally, the user can populate the second Y axis with additional series of data. The result will be displayed as stacked column charts. To proceed with this additional axis configuration, click on the second Axis Options button and perform the same configuration steps as the primary Y axis configuration.

The user can turn on/off these series by checking/unchecking the Hide Primary Y axis or Hide Secondary Y axis.