Monitoring overall performance of assets in federated API management

Monitor and track key performance indicators (KPIs) and business metrics of all runtimes, data planes, APIs in a unified view. You can identify the most and least performing data planes, runtimes, and APIs.

The following key performance indicators (KPIs) and business metrics are monitored on the Dashboard page.

Total number of transactions: The metric counts the total number of transactions that runtimes manage within the landscape.
Average response time: The metric calculates the average time that is needed to process an incoming API request and deliver the response to the client. It includes both the webMethods API Gateway and backend service processing times.
Average latency: The metric measures the average time that is taken by the webMethods API Gateway to process an API request, excluding the time spent by the backend service.
Error rate: The metric quantifies the error rate for failed transactions within the landscape.
Status: The metric indicates the availability status of runtimes and data planes within the landscape.
Capacity: The metric calculates the total number of transactions that a runtime and data plane can handle per second.

The widgets on the dashboard are rendered based on these KPIs. The dashboard retrieves and presents comprehensive analytics for the selected period. It renders based on the following user roles.

API platform provider
API product manager

The following table displays the various sections and corresponding widgets that appear on the dashboard for each user role.

Sections	Widgets in API platform provider view	Widgets in API product manager view
Geographic Visualization	-	-
Count	-	-
Overall view	Total transactions Average response time Average latency Error rate Activities	Total transactions Average response time Average latency Error rate Activities
Status, performance, & used capacity	Status Performance Used capacity	-
Trending runtime & APIs / Trending APIs	Runtime: Upward trending / improving by Runtime: Downward trending / deteriorating by APIs: Upward trending / improving by APIs: Downward trending / deteriorating by	APIs: Upward trending / improving by APIs: Downward trending / deteriorating by
API activity	-	Latest used Most used
Top 5 APIs, runtime, & data planes based on / Top 5 APIs by	Problematic APIs Number of transactions Average response time Average latency Availability Error rate	Problematic APIs Number of transactions Average latency Average response time

The entire dashboard’s content is determined by the chosen time frame. By default, the Last 1 Hour time range is selected. Also, you can tailor the time range to your specific need by using the time-range menu.

The widgets on the Dashboard page are explained in detail as follows.

Geographic visualization

The widget showcases a world map with distinct color pins that visually represents the data planes and runtimes within the landscape. The location of a runtime refers to a city or region for which the latitude and longitude coordinates can be determined by the pin on the map. If the runtimes associated with a data plane are situated in different regions, the data plane is marked at the midpoint of those runtimes on the map. You can use the Search text area to look up a specific data plane and runtime by name.

Upon hovering over the data planes and runtimes pins, the information that is displayed is determined by the selected parameter from the View Settings menu. By default, status and capacity details are shown. In cases where multiple assets are situated in the same region or city, they are clustered as a single group, and the map displays the number of assets within that cluster. Hovering over the cluster reveals the count of data planes and runtimes that are associated with it. To view more detailed information, double-click the individual pins or clusters.

Note: You can drag overlapping pins of multiple runtimes or dataplanes that are located in the same area to improve visibility. It does not modify the original locations of the runtimes or dataplanes. The new pin positions are saved as preferences and remain consistent across user sessions until the pins are modified again.

The Location not assigned widget displays

Data planes where none of the runtimes within a data plane is assigned to any location.
Runtimes that are not associated with any data plane and assigned to any location.
Runtimes with locations that are ambiguous or not uniquely identifiable. For example, Virginia can belong to multiple cities, such as United States | Virginia | Fairfax or United States | Virginia | Virginia Beach. In such cases, you can update the runtime details by selecting the correct country, state, and city from the UI, or by specifying the location in the country|state|city format in the agent configuration.

For more information about how to determine the status, capacity, transaction, and error rate of the data plane, see Metrics of data planes.
For more information about how to determine the status, capacity, transaction, latency, error rate, and response time of the runtime, see Metrics of runtimes.

Count

The widget displays the total number of data planes, runtimes, and APIs within the landscape.

In the API platform provider view, the widget displays the overall count of data planes, runtimes, and APIs in their landscape. In addition, it displays the counts for in-use and unused data planes, as well as active and inactive runtimes and APIs within their landscape.
In the API platform manager view, the widget provides details on the total number of APIs present in the landscape, along with the count of active and inactive APIs.

Overall view

The widget displays the consolidated transaction count, average response time, average latency, and error rate count of all the data planes and runtimes within the landscape. It also displays the activities that are performed in the landscape.

Total transactions: Displays the total number of API transactions that are performed during the selected period. It also displays the transaction rate by comparing the transactions that are made for the selected period with respect to the previous period. For example, if you select the time range as 24 hours, the transaction rate is calculated by comparing the transaction number of the last 24 hours with the previous 24 hours. If the transaction rate for the selected period is increased when compared to the previous period, then it is denoted by an upward arrow else by a downward arrow. Click the total transaction count, the Total transactions pop-up window displays the absolute transactions that are performed by each of the data plane, runtime, or API. Click the respective toggle button to view its absolute transaction value.
Average response time: Displays the average response time of the transactions for the selected period. It also displays the average response rate by comparing the average response time for the selected period with respect to the previous period. If the response rate is increased, then it is denoted by an upward arrow else by a downward arrow. To view the absolute average response rate of the data plane, runtime, or API, click the average response time, the Total average response time pop-up window appears. Click the respective tab to view its absolute response rate.
Average latency: Displays the time that is taken by the runtime to process the request of the transactions made for the selected period. It also displays the latency rate by comparing the average latency time for the selected period with respect to the previous period. If the latency rate is increased, then it is denoted by an upward arrow else by a downward arrow. To view the absolute average latency rate of the data plane, runtime, or API, click the average latency, the Total average latency pop-up window appears. Click the respective tab to view its absolute latency rate.
Error rate: Displays the error rate of the failed transactions for the selected period. It also displays the error rate by comparing the error rate for the selected period with respect to the previous period. If the error rate is increased, then it is denoted by an upward arrow else by a downward arrow. To view the absolute error rate of the data plane, runtime, or API, click the overall error rate, the Error rate pop-up window appears. Click the respective tab to view its absolute error rate.
Note: If there is no adequate data to compute the trend of a runtime, API, or data plane, then the Trend is not applicable message appears. For example, if you register a runtime an hour ago and you try to view the dashboard by applying the Last 24 hours filter, the Trend is not applicable message appears as there is no metrics that are collected for the previous 24 hours.
Activities: Displays the most recent activities that are performed in federated API management for the selected period. Click View all to view all the activities performed in federated API management.

Status, performance, and used capacity

The widget displays the following sections.

Status

Displays the current and previous health status of the runtime or data plane. It is determined by the heartbeat of the runtime. Click the respective toggle button to view the status details. Click View all to view the status of all the runtime or data plane in the landscape.

Performance

Displays the performance of the runtime and data plane. The performance is determined by latency and error rate. The threshold for the latency and error rate is set based on the user preference. Set your preferences by customizing the display settings of the widgets and threshold values for the parameters that control the monitoring capability of the assets. Click the respective toggle button to view the performance details. Click View all to view details of all the runtime or data plane in the landscape.

Capacity

Displays the capacity rate of the runtime and data plane. Capacity is the number of transactions that a runtime can handle. Click the respective toggle button to view the respective capacity rates. Click View all to view the capacity rate of all the runtime or data plane in the landscape.

For more information about how to determine the status, performance, and capacity for the data plane, see Metrics of data planes.
For more information about how to determine the status, performance, and capacity for the runtime, see Metrics of runtimes.

Trending runtimes and APIs or Trending APIs

The widget provides insights into the performance trends of runtimes and APIs, whether they are on an upward or downward trajectory. With the widget, you can compare current and past key performance indicators (KPIs) such as transactions, error rates, response times, and latency, based on the chosen filter criteria. It helps in identifying changes and making well-informed business decisions, which predicts the future performance. The widget highlights the runtimes or APIs based on the actual Trend% value and the trend threshold that are defined in the user preference. Set your preferences by customizing the display settings of the widgets and threshold values for the parameters that control the monitoring capability of the assets. For more information about how to calculate the actual trend% for the runtimes, see Metrics of runtimes. Similarly, calculate the actual trend% for the APIs.

The Upward trending / improving by widgets consist of the following tabs, each displaying the defined upward trending threshold.

No.of transaction: Lists the runtimes or APIs when the actual Trend% value is greater than the set trend threshold.
Error rate: Lists the runtimes or APIs when the actual Trend% value is lesser than the set trend threshold.
Response time: Lists the runtimes or APIs when the actual Trend% value is lesser than the set trend threshold.
Latency: Lists the runtimes or APIs when the actual Trend% value is lesser than the set trend threshold.

Similarly, the Downward trending / deteriorating by widgets consist of the following tabs, each displaying the defined downward trending threshold.

No.of transaction: Lists the runtimes or APIs when the actual Trend% value is lesser than the set trend threshold.
Error rate: Lists the runtimes or APIs when the actual Trend% value is greater than the set trend threshold.
Response time: Lists the runtimes or APIs when the actual Trend% value is greater than the set trend threshold.
Latency: Lists the runtimes or APIs when the actual Trend% value is greater than the set trend threshold.

Example

As an example, assume that the upward trending threshold and downward trending threshold are both set to 10% for all parameters in the User preferences > Threshold page. For upward trending, the set trend threshold is interpreted as +ve 10% for transactions, while it is -ve 10% for error rate, response time, and latency. The following scenarios apply in the Upward trending / improving by widgets.

No.of transaction: Highlights runtimes or APIs when the actual Trend% is greater than 10%. The actual trend% value is displayed in green with an upward arrow.
Error rate: Highlights runtimes or APIs when the actual Trend% is lesser than -10%. The actual trend% value is displayed in green with an upward arrow.
Response time: Highlights runtimes or APIs when the actual Trend% is lesser than -10%. The actual trend% value is displayed in green with an upward arrow.
Latency: Highlights runtimes or APIs when the actual Trend% is lesser than -10%. The actual trend% value is displayed in green with an upward arrow.

For downward trending, the set trend threshold is interpreted as a -ve 10% for transactions, while it is +ve 10% for error rate, response time, and latency.

No.of transaction: Highlights runtimes or APIs when the actual Trend% is lesser than -10%. The actual trend% value is displayed in red with a downward arrow.
Error rate: Highlights runtimes or APIs when the actual Trend% is greater than 10%. The actual trend% value is displayed in red with a downward arrow.
Response time: Highlights runtimes or APIs when the actual Trend% is greater than 10%. The actual trend% value is displayed in red with a downward arrow.
Latency: Highlights runtimes or APIs when the actual Trend% is greater than 10%. The actual trend% value is displayed in red with a downward arrow.

API activity

The widget displays the most used and least used APIs in the landscape. Here, as an API product manager you can analyse the APIs that are least used and make informed business decisions to increase consumption of the APIs.

Top APIs, runtimes, and data planes

The widget displays the most trending items based on your user role. For the API product manager user role, it displays the most trending transactions at the API level. For the API platform provider user role, it displays the most trending transactions at API, runtime, and data plane levels.

Here, you can analyse the most problematic and performing APIs. Based on the analysis, you can derive new business strategies.

You can customize the trending count for APIs, runtimes, and data planes. Set your preferences by customizing the display settings of the widgets and threshold values for the parameters that control the monitoring capability of the assets.

Click the Show as table icon to view the line graph in table format. Click ellipsis icon to download the table in CSV, PNG, or JPG formats.