Earlier this year, we took another step toward ensuring stability and performance in the face of growth (and more growth) by formalizing a DNS Test Engineering department. DynECT Managed DNS offers a high-performance service coupled with reliably constant uptime. We have always tested our products, of course, as we wouldn’t have gotten this far if we hadn’t.
But Dyn’s goal is not to be just adequate but to be leader class.
If we are to lead the way in Managed DNS and Infrastructure as a Service (IaaS), then we must lead the way in reliability, performance, functionality and stability.
This should all be the norm when talking about Managed DNS services, but believe it or not, some organizations find the occasional outage or weak management abilities to be acceptable. Not Dyn. Not for a leader class company.
So how do we plan to test DynECT Managed DNS here in the 21st century?
We use advanced Test Case Management software to coordinate release coverage and manage automated tests. We can develop specific tests for each feature set and variation and run them automatically whenever we want, such as against nightly builds.
Of course, there is also ad hoc requirement testing, as well as testing against multiple configuration scenarios. We have our software tightly integrated with our engineering issue processing system, so development engineers know the instant a test fails and can work to correct the problem.
An obvious key objective for stability is to do all we can to make sure that when new product features are implemented, they don’t have a negative effect on the existing features.
But wait, there’s more.
We want DNS Test Engineering and our development engineers to be tightly coupled, so a different development engineer will work with the DNS Test Engineering department; the engineers will rotate out every few weeks.
This helps insure we have current, valid tests written in part by the engineers themselves and the engineers also bring with them ideas on how to better integrate and improve the testing department and its integration with Engineering. As Subject Matter Experts for their corner of the product, they are also in a unique position to develop specific drill-down testing to fully flesh out our test suites.
The scope of DNS Test Engineering expands beyond merely testing, though. It is considered a key part of an overall integrated software development process.
This means being involved from the very beginning in product and feature proposals, mapping the product architecture both in terms of functionality and implementation, identifying Subject Matter Experts or creating them where needed and working to create a history so there’s a constant flow of knowledge. All these things help both DNS Test Engineering AND the company as well.
They’re mutually beneficial. Everything feeds into each other. What are comprehensive test steps and screen shots today will become the basis for a tutorial tomorrow.
Dyn is growing and understands that with tremendous growth comes a risk that things might be left by the wayside. “Brain drain” or lack of knowledge transfer affects the ability to succeed as people move in and out of positions. Losing control of the development history of products can potentially create a monolith architecture no one understands any more.
Disconnect between testing and engineering risks having a shoddy product.
Dyn is acutely aware that all these things and more can sink a company. One can literally suffocate under the weight of its own success and we are determined for that not to happen at Dyn. We are facing these things ahead of time and being assertively proactive about what needs to be done to make it to the next stage. Our clients expect rock-solid stability and performance and we aim to deliver it and continue being one of the few companies for you that is truly leader class.
That’s DNS Test Engineering.