Original URL: http://www.theregister.co.uk/2014/03/20/gartners_high_end_array_report/

Gartner: Array makers. Think performance counts? WRONG

Crowns HDS array 'best general purpose' high-end kit

By Chris Mellor

Posted in Storage, 20th March 2014 08:35 GMT

A Gartner report looking at high-end enterprise storage arrays published this month has concluded that because of SSDs, an array's performance is no longer a differentiator in its own right.

“Critical Capabilities for General-Purpose, High-End Storage Array” was published on 7 March. It put HDS kit out front for "Best Overall", while VMax kit got an "outstanding" nod.

The analysts looked at 12 products and how they suited five use-cases. The summarised key findings are:

The analyst makes the point that “Product advantages are usually short-lived and temporary in nature.”

There are five use cases:

Lastly there is also an overall rating, giving six measures for the 12 products: Within each of these six categories there percentage contributions for manageability, RAS, performance, snapshot and replication, scalability, ecosystem, multi-tenancy and security, and storage efficiency.

Yes, Oracle’s Pillar Axiom 600 is included, Gartner rating it powerful and functional enough to be in the list, but there is no mention of Dell’s Compellent product.

Gartner’s analysts produced an overall chart for each product based on a score for each critical capability. We reproduced a summary of this using a ruler and photocopy of the report and not the individual product critical capability percentages - that would have been far too much work.

Gartner High-end Array Chart

Gartner high-end enterprise array overall rating chart.

The analyst provides an overall qualitative product viability assessment:

This is Gartner’s assessment of a vendor’s strategy and its ability to enhance and support a product over its expected lifecycle – there was a pat on the back for EMC here.

Then the analyst provides ratings for the overall use case and the five individual uses cases. In each category HDS’ VSP was top with the HP P9000 having an identical score. The measured differences between the 12 products was slight, with a fit to use case score running from 0 (worst) to 5 (best).

Typically the best arrays would score around 4.2 and the worst 3.5, not a great difference at all. Individual products varied position a little in the upper, middle and lower order rankings but the top two positions were settled; HDS VSP and Hitachi OEM'd VSP rules okay.

Here’s the analytics use case ranking as an example:

For more details, Gartner's research note, ref G00248908, is on sale. We scored our copy came from, you won’t be surprised to hear, HDS. ®

Bootnote: The scuttlebutt is that HDS has a new, refreshed VSP coming out in a few months. A scale-out filer type system has also been rumoured to be under development. HDS isn't saying anything about such things.