HP 3PAR kit wins X-Factor for mid-range storage arrays

Mirror, mirror on the wall, which is the fairest mid-range array of them all? Why, the 3PAR 7400, says enterprise data storage analyst DCIG. The outfit rates storage gear in various categories, like enterprise midrange arrays and backup appliances. It updates its reports roughly once a year and is quite successful in getting …

[Nate Amsden]

consistently inconsistent

They seem to list the max cache as a total of the system as a whole, but the max cores of a single controller. Unless it's true that a 16-controller PS6500 has only one CPU core.

Since this article is about 3PAR it would of been nice perhaps if HP had corrected these folks in the RAID levels supported, I suppose perhaps RAID 0,1,5,6 is easy to understand(it is in fact what their own GUI refers to), but reality is of course it's more 0,10,50,60 -- dropping the 0 (from 10/50/60) would actually involve a significant amount of manual configuration that really nobody should ever ever ever feel the need to do).

Also one non trivial error perhaps

Part of their requirements for inclusion is

"It had to scale to support at least 60 TBs of raw capacity. " (bottom of page 6 right column)

(I assume that may be why it seems none of the new fancy flash startup folks are in the list?)

So jump to page 67 "X-IO Hyper ISE 7-Series"

Raw storage capacity (Max)

Should be at least 60TB right?????????????

Maybe it's a typo because it says 36TB

(Same goes for the other X-IO system - just happened to notice that didn't check many other systems)

[Nate Amsden]

and apparently it takes NetApp

24 controllers (12 systems) to do 2.8PB of data on a FAS3250

(this time I guess it is inconsistently inconsistent being that apparently 2.8PB is the capacity of a single system with two controllers yet they show 24 controllers as the max scale out)

ok enough of this it's late I should go to bed

[Lusty]

Ah just like school

If you have enough awards then every kid gets a medal. I'm surprised HP don't have one for best enterprise array with yellow box too.

[Anonymous Coward]

Interesting that the actual PDF is hosted on HP's web site. The document also claims features missing for many competitors where they are actually supported.

It also gives no indication as to why some models get a better "support" rating than others.

I smell a rat.

[Anonymous Coward]

No rats needed

Just like all the other analysts DCIG do their own research and scoring, they then hawk the report to all of the vendors with very likely a high price for exclusivity, HP being the top scorer on this particular one purchased it and so it's hosted for free from their website. All of the vendors have similar relationships with the analysts and where they don't the analysts sell similar reports direct to Customer IT managers.

[M. B.]

Lots of...

...bad data in this report. In a cursory examination I found a number of things missing or improperly tabulated from a number of vendors (support scores being different but the metrics are the same? missing features that ARE supported?). DCIG did a terrible job with this report. As someone with no stake in any of the vendors (customer) I don't think I will be using any of their analysis for my decision-making anytime soon. I've already harshly criticized it publicly on LinkedIn so not going to get into detail here but the report is worthless.

Anyone hoping to use this to assist in decision-making should buck up and do a bake-off instead and get something of value out of their time instead of a sponsored report that's so poorly done I question the integrity of the testers/analysts.

[Anonymous Coward]

IBM XIV Gen3 is very missleading in this report

The number of errors on page 48 is huge.

A couple of examples:

All XIVs have supported (actually required) data redistribution (restripe) when adding additional modules.

The amount of cache memory is off by almost a factor of 8 (Gen3 with 4TB drives has 720GB of memory, this listed 96GB).

Anonymous, as I work for IBM...

[Anonymous Coward]

Re: IBM XIV Gen3 is very missleading in this report

Yeah but then again XIV got additional points for "Mixed Drive Types In LUN" and "Concurrent HDD Mix

In Controller" neither of which it really supports.

The biggest surprise I think was that EMC VNX did so well, but then again VNX has always had lots of features, it's just the inability to use them at the same time that catches out analysts and customers alike.

[Man Mountain]

It's not really saying anything that people don't already know ... StoreServ is the 'best' midrange array. And the reason is that it's an enterprise array masquerading as a midrange array, at midrange pricing. All the other arrays listed are classically midrange.

Even IBM don't seem to be pushing XIV that hard these days, they are all V7000 at the moment. It does change all the time though so maybe XIV will be flavour of the month again soon,.

[rch]

I would consider HUS-VM as midrange+ also since it is a scaled down VSP.

[Anonymous Coward]

Me too

But I do still think the 3PAR has a big advantage having up to four Symmetric A/A controllers in the midrange vs HUS VM's two. Outside of those two and XIV, pretty much everything else in the midrange is active passive even if they do claim ALUA.

[StorageFan]

Is anyone surprised they used a VNX 5700 and 7500?

Those arrays are 3 years old, and have been replaced by the VNX 5600 and VNX 7800...as of 3 months ago.

[Anonymous Coward]

Is anyone surprised

Not at all, but I am very very surprised how well VNX did vs it's competition. If you's read the report VNX2 wasn't available at the time the report was collated.

But lets be honest VNX2 is just more of the same with the latest Intel CPU's and some Lotus go faster stripes, laughably EMC repeat this same launch every 3 years or so. All the same old issues remain but with some magic pixie dust from marketing to make it easy to swallow yet another Clariion / Celerra relaunch. Maybe the XtremeIO product can get them out of that architectural dead end that is VNX.

[StorageFan]

Re: Is anyone surprised

Hah, the marketing pixie dust is too much. They'd be better served to give away a few VNX2s to new customers and write up a story about it

I don't mean to sound too flip - but what "same old issues" are you talking about? LUN tresspassing, etc?

[HPStorageGuy]

A few things to clarify:

> HP didn't sponsor the DCIG report - as pointed out, we did license it and put it on hp.com

> I asked DCIG about the process they used. Participating vendors were given the opportunity to provide input on all aspects of the guide, include IBM. Some vendors provided timely updates per the request from DCIG, others were a little late, still a few more were really late or didn't respond.

> DCIG picked the date that the report would be finalized. Products that were announced after that date weren't included. Simple as that - no conspiracy on why the latest VNX 2 wasn't included.

DCIG told me they also have an online version of the guide that has 150 arrays - that is only accessible by their clients. They keep that information up-to-date.

Bottom line is that IBM has only themselves to blame. Maybe they should go talk to the folks that didn't provide the timely input instead of attacking the report.

Hopefully vendors that are included in future guides will respond with input instead of claiming "foul" after they missed deadlines.

And yes, I work for HP Storage.