How-to: Real-time API Monitoring

Knowing how your API is performing has always been key, but as our APIs become a more important and a key part of your customer experience it needs to be looked at differently. Traditional API monitoring relies on the API and its components as part of application infrastructure. Although this is important it also misses the most important KPI for an API, the performance and availability of the API from the consumers perspective. This could be your mobile application, website or B2B channel, how they see your API is how it is judged. Your API is a product and monitoring it from the user’s perspective is the best way to understand your customers experience.

During the current COVID-19 crisis we are seeing more Layer7 API management customers experience a move to their digital channels. With stores closed and face to face meetings impossible the demand on APIs has increased massively for nearly all organizations. The obvious, headline makers, like ZOOM and Netflix of course have demand increasing,as we try to stay entertained and in contact with our personal and professional networks. But all organizations from retail to telecoms and importantly healthcare are seeing ramped up demand in their API usage. We are all learning to work, play and look after ourselves in new ways.

Of course,topics like scaling and availability are very important and we will look at those in another blog soon, but first lets looks at how your API is performing from the user perspective. API monitoring from the cloud can give you a massive advantage in judging how you are being judged. With BlazeMeter API monitoring you can make real calls to your API from a cloud-based solution, may be even running test from multiple locations to get a wider picture of how networks and geography effect your API. The BlazeMeter API monitoring solution can give you the simple response health of your API but is an HTTP 200 enough. The solution should also be able monitor performance and inspect the content being returned to the client. A 200-response code can show that the HTTP call worked but being able to see and interrogate the content is key to identifying issues with your API and the infrastructure behind it.A payload of irrelevant data is as bad as no data!As more and more microservices development teams subscribe to the ‘you build it, you own it’ model, they create and run the necessary tests to ensure their APIs are secured, work and perform as expected in pre-production. In addition, they want to reuse those pre-production API tests in production (a.k.a. “shift-right” -shift testing to the right of the SDLC) so they can get immediate alerts when their API is not working or performing properly in production. They can then react and take instant action (e.g. roll back/forward) to prevent impact to the user experience.

In this short video we will look at how we can use BlazeMeter API Monitoring to give us the customer view of APIs being managed in Layer 7 API Management.