API Management / Development / Sponsored / Contributed

How to Test for API Reachability in Your Apps

9 Feb 2021 9:24am, by

Joe Dougherty
Joe Dougherty is a product manager, leading the development of the Browser Synthetics product at ThousandEyes, a Cisco company delivering cloud and internet intelligence.

Modern applications are a bit like icebergs. The user sees the top — the point where interaction or transaction happens — and draws their conclusions from there about the business. Ideally, it’s a seamless, simple experience.

Like an iceberg, though, what the user sees is only a small part of the whole.

Most applications today, even for something as small as placing an item in your cart, rely on a vast web of services and platforms distributed across the internet — typically via a REST-based API.

That might mean a messaging service that uses Twilio for cloud communication, an e-commerce site that uses Stripe for payment processing, or a delivery service that uses Google Maps for geolocation — among countless other examples. To optimize the delivery of these digital experiences, we need to understand how the APIs are performing. When it comes to ensuring overall application performance, understanding API reachability over the internet and cloud provider networks is crucial.

Lack of Visibility Makes Solving Issues a Challenge

When problems arise in the application workflow, you need a fast fix. But the complexity and distributed nature of modern application workflows make finding the root of the problem difficult and time-consuming. Not good when your reputation rides on your app’s performance.

Legacy network and application monitoring tools, while useful, don’t provide the level of visibility required to quickly understand the problem, escalate appropriately, and resolve issues across external workflows. Packet capture and flow analyzers, for instance, don’t work outside of your own environment.

This means businesses are often blind to most of the delivery path, and therefore the source of most issues that impact user experience. And to properly escalate and resolve a problem outside of your own infrastructure, you need enough evidence to convince third parties to take action.

Without that evidence, you’re left spending precious time and money on help desks and inefficient troubleshooting, while your users are left with a problematic application experience.

We’ve found at ThousandEyes that there are other challenges beyond simple visibility. Delivery paths are constantly in flux. There is no steady-state in the cloud. For example, if you’re dependent on a third-party API based in Ireland, there’s no guarantee that it will still be based there tomorrow. Data centers pop up, relocate, or disappear entirely. All this can directly impact the functionality of your application. This means we need more and better tools to properly address issues that arise.

While browser synthetics offer a powerful tool to test applications and measure user experience across entire workflows, there are instances when a single request in the browser triggers multiple backend API interactions — which cannot be directly observed from the user’s perspective. For example, when a user submits an order form on an e-commerce website, the application makes a series of API calls to check inventory, process payment, and generate an order number — before directing the user to an order confirmation page. Because these backend services are invisible to the client, a failure or performance issue in any one will ultimately go undetected and directly impact the customer.

Instead, application owners need a testing approach that doesn’t rely on front-end interaction only, but rather incorporates testing from within the context of the core application in order to understand the impact of underlying network transport — typically Internet Service Provider or cloud provider networks. This is where adaptive API monitoring comes in.

Adaptive API Monitoring Offers a Way Forward

Adaptive API monitoring goes a step further. By executing sequential, conditional, or iterative API calls against your API dependencies, it offers a highly flexible synthetic testing framework that emulates backend application interactions with remote API endpoints.

API monitoring tests can be run from vantage points that are external to the application environment, or from agents placed within the application-hosting environment out to the API services. An advantage of this latter deployment approach is that the specific network paths and performance from the application to the API endpoints can also be monitored.

All this adds up to powerful testing options and unprecedented visibility across the external networks that impact application experiences. For application owners, this means you can dynamically measure performance, differentiate timings between each iterative function, and validate the logic of complex workflows. The benefit? Rapid problem solving, smarter workflows, and valuable insights into opportunities for optimizing the digital experience.

Feature image via Pixabay.

The New Stack is a wholly owned subsidiary of Insight Partners. TNS owner Insight Partners is an investor in the following companies: Bit.

A newsletter digest of the week’s most important stories & analyses.