By: John S.

John S. — Wed, 07 Dec 2022 14:55:22 +0000

I think part of the problem with today vs back then is that in it’s day Netapp was targetting a much broader market, as you alude to. And EMC was targetting the C-level with the name “BCVs” for it’s marketing. Hard to argue with “Business Continuity Volumes”. But today new sales are are either much smaller in scale, or they are competing with the cloud which is all about large large large scale. So there’s very little in the middle any more. Which is why Netapp is working so hard to get into the Cloud and to be the on-ramp for their legacy on-premiss storage into a cloud setup where the costs (up front, not long term) are much lower.

I’ve been using Netapps since 1996 and I really like what they do. They’re rock solid, have tons of features and really do a great job. But they’re also fighting the 10% problem, where 10% of their customers want 20% of the features, but each 10% wants a different subset. MetroClusters are really key for a very small subset, but those people drive a bunch of revenue. But is it worth it? Maybe.

And once you get the HPC space, the costs of large Netapp clusters is huge, and at that point maybe it makes more sense to run Lustre/Ceph/GFS since you have so much scale and it’s trivial to get reliability and performance by cheaper scaling out of the base product, instead of paying the Netapp/EMC/Pure/Nutanix tax over and over again.

And of course Nutanix isn’t really the same as EMC/Netapp, they’re chasing a different market. It’s hard to justify really reliable systems for the SMB market because people are so cost sensitive and focused on the bottom line. It’s expensive to get past RAID5 clustering or sharding or whatever for the HyperConverged hosting market, since people look at the price of a 2tb SSD for their home system being $300 and wonder why they’re paying $3000 for the same thing in usable space.

Preaching to the choir I am.

Comments on: Can Anyone Make Money From Modern Storage?

By: John S.