AdroitLogic IPS, Part 1: What it is (and Why You Should Care)
Knowing the yet-again-proven performance margin of UltraESB-X, you may perhaps be wondering what it takes for a brood of them to run in unison, load-balancing and gracefully handling your business demands (web traffic, for instance). Well, never fear, because IPS is here!
Okay, I won’t lie: IPS has been around for quite some time (in fact, since last February), though we didn’t give much of a fuss about it. But now, as we continue to make the platform more resilient, feature-rich and user-friendly, we have decided that the world has a right to know.
While IPS was initially written for Kubernetes, later it was expanded to support OpenShift as well; we even came up with an AWS ECS-based version. Right now we are working on something bigger, a cloud-hosted version; more on that later.
IPS can spawn a cluster of ESB instances, all with identical runtimes, across a batch of worker nodes. Once any services running inside are exposed for external access, IPS automatically starts load-balancing them across the spawned instances. The health of the cluster is taken care of by IPS, with immediate and automatic respawning to balance out any dead instances, and thorough monitoring and (optionally) automatic rollback of failed deployments.
As if that wasn’t enough, IPS gathers the logs and statistics published by each ESB, and exposes them via a comprehensive dashboard, with advanced search, filtering and drill-down capabilities. It also exposes platform-level health via real-time logs and health monitors. A cool snapshot feature allows you to capture the running state of an ESB instance — or even a whole cluster, complete with its current deployment configuration — any time you wish.
On top of all this, we added two new hot features in the latest 17.07 release. One allows you to start and stop individual ESBs and deployed projects in a cluster without having to go for a redeployment, and the other introduces detailed monitor logs that track and display status changes for your cluster, with detailed downloadable diagnostic information whenever a failure is detected. In addition, numerous enhancements have been done to the base framework (not to mention UltraESB-X itself, so that IPS is now more stable than ever.
Ready to take up a challenge?
Head over to the next step: get your own demo up and running, on your own machine!