solo.io Gloo Platform, 7 Jul 2023

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Gloo Platform Portal Analytics and Monetization Demo

Description

Welcome to a demonstration of the powerful analytics feature in Gloo Platform Portal API. In this video, we'll showcase how this feature provides valuable insights into API usage.

Our platform's analytics workflow begins when a user or system accesses an API using an API key. The API Gateway, which is an Envoy proxy, validates the API key and verifies authorization for API access. During this validation process, we retrieve additional metadata related to the API key, such as the attached usage plan. This information is then stored in the access log of our API Gateway.

To harness the power of analytics, we employ an open telemetry pipeline to fetch data from the access log. This data is processed within our OTEL (OpenTelemetry) pipeline and stored in one of the supported data stores. In this demonstration, we utilize ClickHouse as the data store for the API access log. The stored data includes requests, response validity, API product access, specific API access details, usage plans, API keys, and more.

By leveraging ClickHouse as a data source, we can visualize the API usage information through a Grafana dashboard. This dynamic dashboard allows us to select the desired product and view usage information based on various parameters. We can choose the usage plan, request URL, HTTP method, response status code, and even filter by user ID. This enables us to analyze request latency, track status codes over time, and identify top API consumers for a specific API.

Furthermore, the dashboard allows us to focus on individual API users by selecting their user IDs. We can examine up to 100 user IDs, matching the top API consumers. Additionally, if we wish to explore the usage of a specific user, we can easily apply a filter based on the username or a partial match. This provides comprehensive insights into total API calls made, status codes over time, request latency, and more.

Apart from the overall API product usage and analytics dashboard, Gloo Platform Portal offers out-of-the-box dashboards for each API product. These dedicated dashboards provide focused information for specific API products or allow users access to relevant analytics for their respective products.

It's worth mentioning that our open telemetry pipeline allows data export to any data store supported by OpenTelemetry. This means the data utilized in our API product usage and analytics dashboard can be exported to other environments for further processing. This opens up exciting possibilities for implementing use cases like API usage monetization and advanced analytics.

Prepare to dive into a world of actionable insights as we explore the API usage analytics capabilities of Gloo Platform Portal. Stay tuned for more features and demonstrations!

A

In this video, we will demonstrate the gloo platform portal api uses an analytics feature.

A

This diagram shows a high level overview of this feature of our platform. when a user or.

A

System accesses an api using an api key or api gateway,, which is an envoy proxy in our case,.

A

Validates the api, key, and checks whether the user or system is authorized to access the api.

A

During that validation, we retrieve additional metadata for that api key,. For example, the usage.

A

Plan that is attached to that api key, and we store that information in the access log of our.

A

Api gateway, and using an open, telemetry pipeline,. We fetch the information from that access log and.

A

We process that in our otel pipeline to later be stored in one of the supported data, stores.

A

So in this example, we're using clickhouse as the data store for the api access log. So.

A

That contains the request,, but also whether the requests will be valid,.

A

Whether they got a correct response or an error,, the api product, access,.

A

The specific api access,, the usage plan,, the api key and so forth.

A

From clickhouse,, we can then use that as a data source for a grafana.

A

Dashboard, which allows us to create a view and to show a dashboard of the.

A

Api usage for the products and the api products deployed on the portal.

A

Here we see the grafana api dashboard, and in this dashboard we can select the product for which we.

A

Want to view the information. we can select the usage plan.. We can also select the request. Url,.

A

So the actual path within the api product that we're interested in., the http method, the status.

A

Code of that request in response, and the user id for which we want to see the information.

A

So when we select a specific api product,, we can see that the information that we.

A

Retrieve is the information for that specific product.. We can check the.

A

Request latency of the request to that api product.. We can see the status code over.

A

Time as well as things like, the top api consumers for this specific api.

A

So, let's switch this back to all api products.. Another very useful feature is that we can.

A

Select the api users per user id.. In this dashboard we can show up to 100 user ids here,.

A

Which match the user's id of the top api consumers,? But if you want to look.

A

A specific user and their api usage,, you can simply use that filter and type in.

A

The username or part of the username,, and then we can filter on that specific user.

A

And then for that user we can see things as the total api calls that they made,.

A

Status code over time of the requests and responses, and the request, latency and so forth.

A

Besides the overall api product usage and analytics dashboard,.

A

We also provide out-of-the-box dashboards per api product,, so if you are only interested in.

A

The information of a specific api product, or if you want to give users access to only.

A

The information of a given api product,, those dashboards, are readily available for you to use.

A

Finally, we want to point out that, because we use an open, telemetry, pipeline,.

A

We are able to export our data to any data store, supported by open telemetry,.

A

And this means that the information that we use in our api products, users and analytics dashboard.

A

Can also be exported to other environments for further processing,, and this opens the.

A

Way to implement use cases for things like api usage, monetization,, analytics, and so forth.