Gartner's Magic Quadrant mages shake crystal ball, Violin goes topsy turvy Gartner's latest all-flash array Magic Quadrant contains expected news, a few surprises and a demotion that has prompted a public response from Violin Memory. The report, Gartner's 2016 Solid State Arrays Magic Quadrant and Critical Capabilities" report, provides a snapshot of vendor positions whilst the Critical Capabilities …
This is an AFA only GMQ so is very specific. Dell don't really have anything exciting in the AFA space which would warrant being on it. They can sell you an array with only flash in it, but it's not really any different.
NetApp had to play this game, and have since innovated on their AFF platform enough to move the right way.
Surprised to see Pure on here where they are, I wouldn't call their vision complete even if they do make nice boxes.
I would love for Gartner to defend this stance. How is Pure's "vision" more complete?
- Do they support the latest most dense SSD? No.
- Do they scale to a PB in one system? No.
- Do they offer advanced replication offerings? No.
- Bit level error detection to guard against silent errors? No.
- How about scale beyond a single controller node's ownership of volumes? No.
- Granular, in depth reporting? No.
- How about the "//m90" that was supposed to be here? Is it? No.
- NVMe? No
Even in the write-up Gartner gave Pure credit for 2 things: "Mindshare' (does anyone know if that's a thing?) & Price point. How does that make their "vision" more complete than any of the other vendors?
And 'mindshare'? Really? That's just bs
Dell's kit does not meet the Gartner's inclusion criteria. There's a number of criteria but my guess it's because Dell don't offer a Solid State only array which cannot be reconfigured, expanded or upgraded at any point with any form of HDD. In other words, if it's flash retro fitted to legacy then it aint no AFA.
If you get 'stuffed' by a Gartner report just find out which one of your competitors paid for the report and you will know in an instant who thinks you are a real threat to them...
As for the Wall St {insert derogatory name of your choice} Anal people who seem to hang on every word in these reports... well they can and do make or break a company.
If I were in a position to do anything about it, I'd send the lot of them to ride the coney Island rollercoaster until they see the error of their ways. If they don't then .... point them at the Atlantic and tell them to swim it with no assistance.
Gartner should merge the AFA MQ to the external disk MQ as the two have converged. Flash/SSD is merely a technology/technique and not a real distinction between storage products.
1. All HDD vendors have become AFA vendors: EMC, Dell, IBM, HP, HDS, Netapp.
2. All hybrid vendors have become AFA vendors: Nimble, Tintri, Tegile.
3. All hyperconverged vendors have become AFA vendors: Nutanix, Simplivity, VSAN.
So why is there a separate MQ for AFA vs. MQ for External disk? Those are the *same* vendors catering to the *same* customers, all vying for the *same* market.
Stated differently, what does EMC XtremIO or Pure Storage displace, if not HDD-based products?
There were good reasons for setting this up as a separate analysis IMO; there is a lot of difference between a traditional array (even one with SSD exclusively in it) and an AFA.
There is a combination of:
1. Optimisation of IO to deal with fast throughput of SSDs/Flash
2. In-line dedupe to reduce cost/GB
3. In-line compression to reduce cost/GB
So an HDS VSP/G1000 etc or EMC VMAX with Flash is not an AFA. A Tintri VMStore or EMC XtremeIO are an AFA because their architecture has been built from the ground up with this in mind.
This is also why we see NetApp move so far over - they changed from their crappy kludge interpretation of an AFA with their FAS and purchased SolidFire.
(I have previously worked at both HDS and Tintri - I no longer work for a storage vendor)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
