The Analytics subsystem is what receives the transaction details from the gateway cluster. After you install the Analytics service, you can then associate the Analytics subsystem with a defined gateway cluster. Once associated, the gateway will begin sending transaction data to the Analytics subsystem:
Figure 2.11 – Associating the Analytics service to a gateway
Following this, the analytics data can be reviewed by the Analytics user interface that is displayed in API Manager and Developer Portal. The Analytics capabilities that come with API Connect are based on the open source products from the Elastic Stack (Elasticsearch, Logstash, and Kibana or ELK). Here, the user interface is using Kibana.
As an API developer, users can view analytics within the Provider organization Catalogs they are associated with. This includes spaces that they have been added to.
You can only customize analytics from the Provider organizations. Whenever you click on the Analytics tab, you are presented with the choice of preconfigured dashboards:
Figure 2.12 – The Analytics tab within a Catalog
Within each dashboard, there is a preconfigured visualization of the metric data. The screenshot that follows shows you the visualization for the default dashboard. You have the capabilities to modify the visualizations and/or create your own:
Figure 2.13 – Visualizations for the dashboard
With Analytics, you can also search for analytics produced from a different period of time. On the right-hand side, you can find the default time period of Last 15 Minutes. As shown in the following screenshot, if you click on that link, you will be given the option to select a different time period:
Figure 2.14 – Changing the time range
You might have noticed a dashboard called Portal default. This is where you can view and modify the analytics that are available for the Developer Portal.
Within the Developer Portal, API consumers can view analytics under two categories (consumer apps and consumer organizations). They can view analytics under the Apps tab for the consumer apps that have been created. The apps are the implementations of the subscribed product and APIs. The analytics that can be viewed include API stats, total calls, and total errors.
The other place in which to view them is based on the consumer organization. The consumer organization is the organization of the developers subscribing to your Product APIs. The information categories are the same (that is, API stats, total calls, and total errors) but rolled up just for that consumer organization.
There is a lot more you can learn about the Analytics subsystem. You can learn how to configure the Analytics dashboards and visualizations by referring to the documentation on the web about the Elastic Stack.
Important Note
While not covered in this section, you should be aware that the analytics data can be exported from the Analytics subsystem. Additionally, you can configure API Connect to send the analytics to another server for analysis and reporting, such as Splunk. If you would like to do so, please refer to the IBM Documentation web page.
So, you have learned about the four major components of API Connect. You learned that Cloud Manager is where you configure the API Cloud topology and link the components. Additionally, you discovered that API Manager is where you create products and APIs. The Developer Portal is the second component where you can publicize your products and APIs as well as allow subscriptions to consumers to develop applications. We introduced the DataPower gateway where your APIs actually run, and we described how analytics are sent from the gateway to the Analytics subsystem for analysis and reporting.
By now, you should be familiar with the various components that comprise an API Connect cloud. Next, you will learn about the deployment models for an API Connect hybrid cloud.