Chris Phillips’ Blog - API, Integration and Governance

“Which portal should I use?” This is one of the most common questions I hear from teams implementing IBM API Connect. The answer isn’t always obvious—IBM offers four distinct portal options, each designed for different use cases. Choose wrong, and you’ll either over-engineer a simple solution or find yourself constrained by limitations you didn’t anticipate.

Let me walk you through the four options, when to use each one, and how to make the right choice for your API program.

Your API Gateway is slow under load. GatewayScript engines are maxed out. Requests are queuing. Sound familiar?

Understanding the relationship between GatewayScript engines, CPU cores, and concurrent execution is crucial for building high-performance API solutions. This guide cuts through the complexity with battle-tested formulas and real-world recommendations for handling high-concurrency scenarios.

Stop hardcoding secrets in your APIs! Learn how to build a reusable custom policy that seamlessly integrates HashiCorp Vault with IBM API Connect, transforming complex authentication and secret retrieval into a single, elegant policy call.

📦 Get the Code: The complete policy code, deployment scripts, and examples are available on GitHub: apiconnect-vault-integration

I know some of you have noticed I’ve been publishing less on my blog lately. The reason? Less travel, thanks to these little monsters. 🦖👶

Image description for accessibility: Photo shows a bearded Chris Phillips wearing glasses and a black t-shirt with a iSOA skull design, sitting with his two young blonde daughters. One daughter wears a light blue dress with colorful dinosaur prints, the other wears a light pink dress with unicorn prints. Both children have pacifiers. The setting appears to be a casual restaurant or café with other people visible in the background.

Thanks to my darling brother Prof Iain Phillips for the great photo of my little girls.

I’ll be the first to admit it: I wasn’t practicing what I preach. For someone who regularly writes about best practices, and keeping systems up to date, I had let my own Jekyll blog fall embarrassingly behind. My Jekyll version was ancient, my Ruby dependencies were outdated, and I had been ignoring the problem for far too long.

This is terrible. Not just because outdated software can have security vulnerabilities, but because it goes against everything I advocate for in my professional work. If I’m going to write about proper software maintenance and infrastructure management, I need to walk the walk.

So today, I decided enough was enough and (with a daughter asleep on top of me) it was time to update my blog’s Jekyll installation and get everything current. What followed was a journey through Ruby version management, dependency resolution, and the occasional head-scratching moment that reminded me why keeping things updated regularly is so much easier than letting them fall behind.

Here’s what I learned (well, re-learned) during the process.