[Originally published on LinkedIn 30 May 2024
Mehdi Daoudi, CEO of Catchpoint talks to us about internet performance monitoring (IPM) at AppDev Field Day 1.
Publicly facing internet applications are part of an extraordinarily complex “internet stack”. Starting at the top of the stack, we have:
- Applications
- Media / Add
- cloud services
- Internet Core
- Protocol
- Network
Why do you need to understand your internet stack?
You really want to focus on the trinity of revenue, brand (risk & compliance), and excellence. These are affected by your user experience, so monitoring user experience issues is paramount. While many organizations have implemented an application performance monitoring (APM) solution to understand the user experience, that user experience goes beyond your application stack to include the entire internet stack
APM vs IPM
Traditional APM is different from internet performance monitoring (IPM). APM monitors what you have control over: your application, your servers, services, and infrastructure. IPM is understanding how the internet is impacting your user experience using real-world interactions from across the globe.
What is Catchpoint’s IPM Products?
- Synthetic monitoring – pretending to be a user and testing DNS, web, api, like “mystery shoppers”.
- Real User monitoring – analytics for real user activity.
- Internet synthetic monitoring – telemetry from synthetic users.
- BGP Monitoring – detect and resolve BGP routing issues.
- Endpoint monitoring – look beyond the device.
- Web page test – test and accelerate web pages.
- Internet sonar – is it me or something else?
- Tracing – enable IT ops teams to quickly fix app issues.
Diving Into Internet Sonar and the Internet Stack Map
Internet Sonar looks at 15M points across the internet to help you understand if there is something going on that will impact your user experience.
Internet sonar drives Catchpoint’s Internet Stack Map, a live view of your service and service dependency health. With this view, you can understand which issues in which part of the internet stack are impacting your app.
Internet Sonar includes information from network providers such as outages and estimated repair times, so that you can avoid wasting your time initiating service calls.
An important part of the solution is the more than forty out-of-the box testing and monitoring types, such as SSL, DNS, BGP, etc. and the flexibility to support an infinite number of bring-your-own monitoring options. This enables you to run instant on-demand tests that matter to you from anywhere around the world.
In Brandon DeLap‘s example use case, Catchpoint shows a set of errors in a graphical map of your internet stack. Issues are highlighted in red, with an icon for issues that have suggested remediations.
Double clicking on a web page error shows you a trace of the transaction to understand where in the stack the error occurred.
Similarly, you can click on a DNS error (because, as the internet meme says, “It’s always DNS!”) Catchpoint will show you a map from Internet Sonar where all the DNS failures occur.
[edited to add this really cool piece that I first missed…]
The secret magic sauce is that Catchpoint’s graph can trace a transaction from the user through the internet to your application and then all the way down to your source code. If that doesn’t help you understand, identify, and mitigate issues, I don’t know what will!
Is Catchpoint Really Useful?
Using Catchpoint, global companies have moved from being reactive to proactive, experiencing massive improvements in key metrics including:
- Time to interact (TTI)
- MTTR
- SLA Violation %
- Total Incidents/Year
- Total Major Incidents/year
- Customer Incidents/Year
When I say massive, I really do mean massive. One company was able to reduce MTTR from 5 hours to 3 minutes!
What about AI?
[edit/update] Gerardo A Dada pointed out that Catchpoint has multiple AI-powered capabilities such as automated experiments, automatic correlation, anomaly detection, or even business impact estimation. It also understands the usefulness of trillions of historical data points for building and training AI models that will be able to detect and predict issues and recommend or automate remediation activities.
Is Catchpoint’s IPM a Good Fit for You?
Catchpoint differentiates between testing and monitoring. While it does have a web page testing product, the focus is on detecting dynamic issues with the goal of helping you reduce mean-time-to-repair (MTTR).
Catchpoint provides you with information about both your application and how it works in the global environment — from the perspective of the end user. The more information you have, the better your decision making and your ability to adjust to provide the best user experience.
While you may think that this service is only about helping you identify if the issue is within or outside of your control, and fixing issues that you do control, Catchpoint goes beyond that.
You might not be able to fix issues outside of your control, but with enough information about these issues, you can make other adjustments to mitigate the issue and improve every one of your user’s experiences across the globe.
And that’s what really matters.