Evan Miller

How do I accelerate the clock cycle of everything we do?

That's much, much more interesting than sitting down on the keyboard and provisioning storage and then getting blamed when everything goes wrong.

Because when humans do everything, it's inevitably prone to mistakes, right?

Whereas if you automate everything, you get extreme standardization.

So the key understanding here is that a disk drive is not a service, right?

SATA, SAS, SSD, different sizes of drives, different vendors, different controller types.

None of those things constitute a service.

An application doesn't consume any of those.

An application does one thing.

It consumes IO at a latency.

And so that has to be the performance promise.

That's fantastic.

And another way that this is completely unique in the industry is it's called adaptive QoS.

That's a fundamental difference between this and every other approach to maintaining a service level.

Number one, we don't want anybody to have to, like, tweak this thing, right?

You put a volume where it needs to be, and it just does what it needs, right?

And if the volume grows or if the amount of data grows, you want to make sure there's enough IO there for the workload, and you don't want to have to have somebody, you know, calculating IOPS anymore.

And so adaptive QoS adapts the QoS policy to the size of the volume and the amount of the data in the volume.

For example, if you grow or auto grow the volume, the IO grows with it.

If you put more and more data in that volume, particularly with compression and dedupe and compaction, all those technologies can sometimes store two to four, sometimes eight times as much data in the volume as the physical footprint.