Jeff Baxter

And, you know, our CEO, who at the time was running ONTAP and then Product Operations, George Kurian, you know, had a great analogy, which the problem is that the train only leaves the station once every couple years.

No matter what, as you start to depart, someone's going to be running up behind the train saying, just wait, let me on, just wait, let me on.

So you pause the train and they jump on.

And then someone else runs up going, you know, just let me on, just let me on.

Whereas now what we've moved to is the train leaves the station every six months.

And there may not be, you know, 15 major features on every release, and that's okay.

There may be more like six or seven major features each release, or maybe only three or four.

Over the same 18-month two-year period, you're going to get more innovation, but it's going to come at more predictable regularity, and it's going to be more sort of bite-sized chunks of innovation.

One of the things our customers would tell us is, you know, it's awesome all these things you have in 8.2, or it's awesome all these things you have in 8.3, but it takes us a year to qualify because there's so many new things in here, and we don't know the impact that will have on our environment.

Instead, with each of these new major releases of ONTAP, they're still qualified, they're still fully onboarded as a major release of ONTAP, but the amount of churn and the amount of change and the amount of new features to integrate into your environment is a lot easier to consume on the sort of six-month cadence model.

Yeah.

I mean, I'm honestly expecting that there will be customers who will deploy every year, right?

They'll skip a release or something along those lines.

We're not anticipating that every customer is going to absorb every release just like they did in the past, but this gives them a little more regular cadence of when things arrive.

And you're right.

both as a factor of the legacy of how we were shipping and the pivot to that.

I mean, moving to six months has been a pretty seismic pivot as it is.

And then the criticality of where we set the stack.

I mean, you can drop a packet or you can have a server that reboots and it could have negative impact.

But if you lose data, it's kind of game over, right?