back to article Gartner: Array makers. Think performance counts? WRONG

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 …


This topic is closed for new posts.
  1. Anonymous Coward
    Anonymous Coward

    These sorts of tests are utterly pointless. If you were to compare all cars currently available, the ones coming out top will probably all be sports cars. All useless if you have three kids.

    Even the title gives this away. "General Purpose, High-End Storage Array". If you're using a high end array for general purpose storage in this day and age then you clearly have money to burn.

    Seems to me like Gartner trying to be relevant. All their reports only ever get used by vendors who select the ones which favour them to use as a marketing tool. Like this article.

    Anyone with any sense whatsoever who has this sort of money to spend will select a shortlist based on merit and cost, and run a proof of concept.

    1. Robert Grant

      Agreed. Was about to write what you did!

    2. Anonymous Coward
      Anonymous Coward

      You didn't even bother to read it, did you?

      IF you had read read the report, you would have seen the descriptions for 5 very common uses cases. Oh well, it's easier to troll than to read.

    3. DougS Silver badge

      I don't think the comparison is unfair

      Typically you'll use a high end array for business critical apps. Some of them may be specialized (SAP or whatever) but even general purpose filesystems containing web pages may be on a high end array if that web server is considered business critical.

      You use less costly midrange storage for apps that don't reach the CEO's attention when they go down.

      1. Don Jefe

        Re: I don't think the comparison is unfair

        The merits, costs and performance of a piece of kit are all secondary to a viable entity to blame for less than stellar results.

        Everyone forgets that market analysis companies are just marketing companies, no different than ClearChannel or Google AdWords. All the costs of dealing with them comes from your marketing budget and it is almost always your marketing people you send to deal with the analysts.

  2. Anonymous Coward
    Anonymous Coward


    News Title includes word Gartner - IGNORED....

    1. Anonymous Coward
      Anonymous Coward

      Re: HA

      Really, it's actually quite an interesting read and, unusally for Gartner documents, free!

      Gartner like anywhere is a bit hit and miss, some of the best experts I've ever talked too, some of the fluffiest, so YMMV. But when you're spending multi-millions on a deal, they're worth a chat even if you wouldn't take their call as gospel truth, but even they say that. I find most of the people who badmouth them tend to be from vendors who don't feature or do poorly then it's all, "pay to appear" and "sponsored entry" without any evidence or a justification of why, if it were true, didn't they just pay and what the price was.

      1. Don Jefe

        Re: HA

        Gartner sending out free reports isn't rare, at all. You may be primarily exposed to the 'not free' and not aware of the free stuff.

        Pricing, or lack thereof is determined by the intended audience of the report. If the report is targeted at end users of the product being discussed the report will be free (for you/end user). The costs are paid by whoever commissioned the report.

        If the report is something you will use in seeking investment money, validating/justifying a strategy or feed to your defense lawyers, or anything else that benefits you directly, the report will have a cost. As you note, that cost is often substantial.

  3. Anonymous Coward
    Anonymous Coward

    Ah, Gartner. The favourite reserve of the management consultant who wants to appear knowledgeable. I cringe whenever a higher up or consultant start referencing Gartner. I know that we'll go through a lot of pain and wasted effort just to make someone feel important or useful.

  4. Anonymous Coward
    Anonymous Coward


  5. Nate Amsden



    One of my friends told me a few months ago about someone he knew that got hired (back) into Pillar as a sales rep, and his account? His account was to sell Pillar to ORACLE (internally of course). How hard can that be? Selling a product you make to your internal groups.. Hard enough apparently because he struggled to even do that! Not even Oracle wants to use Pillar! Sad. Just put it out of it's misery already.

    Pillar certainly should not be on any high end storage list nor should IBM XIV (it's single system scalability has always been a joke - I mean come on max of 180 disk drives in 2014?). Nor should HDS HUS, Huawei, and I'd argue no NetApp is high end, not with their architecture. Sure you can build a "big box" with lots of disk drives but the architecture falls apart compared to things like VSP, or VMAX or 3PAR(and yes that means the NetApp clustering doesn't make the cut either).

    HDS VSP and P9500 certainly do seem to be good high end kit though obviously very complicated systems with an ever shrinking market size. HP has all but stopped marketing the P9500 from what I've seen post 3PAR acquisition and is aggressively going after P9500 customers with 3PAR systems with good success last I heard. 3PAR can't do everything VSP can for sure(and it may never - some of those things there just isn't enough demand for them), though I think the capabilities gap is greater in the reverse(VSP can't do everything 3PAR can do, or at the very least does them very poorly relative to 3PAR).

  6. M. B.


    ...isn't in the list because the others all appear to be able to do true multi-controller configurations (with many caveats in some cases, Live Volume between Compellent heads seems to be nearly as functional as Cluster Mode).

    But the gist of it is the EMC, Hitachi, HP, Huawei, Fujitsu, DDN, and Oracle arrays can all cluster across more than 2 controllers. I wouldn't really consider Oracle or the baby Hitachi considering the others in this competition though. Probably not NetApp either,, for the same reason Compellent doesn't fit.

  7. Anonymous Coward
    Anonymous Coward

    More Gartner pseudo science

    As usual Gatrner try to cover too much ground, from too higher level, whilst including some of their favourites and excluding others. Oracle, Netapp, XIV and HUS-VM shouldn't really be in this list, but no doubt Gartner's Customers expect to see these names, so they have to squeeze them in somehow.

    1. Anonymous Coward
      Anonymous Coward

      Re: More Gartner pseudo science

      Bullocks. XIV and NetApp FAS/V6200 are both in the same class of products and are head-to-head competitors.

      1. Anonymous Coward
        Anonymous Coward

        Re: More Gartner pseudo science

        My point exactly,they may be genuine competitors with each other but not with the rest of that pack. XIV scales to 180 disks, FAS is a dual controller active passive array, unless your running multiple active passives in cluster mode. But even then it requires a large dose of wishful thinking to compare it to the VMAX, VSP or high end 3PAR arrays.

  8. Anonymous Coward
    Anonymous Coward

    EVERY Enterprise IT shop relies on Gartner

    You can make jokes about it but as painful as it is, every large IT shop (and especially their CTO's) relies on Gartner to determine the short list of vendors for evaluation.

  9. Anonymous Coward
    Anonymous Coward


    Wonder about the objectivity of this, and if it is more than evaluating perceived opinions plus studying websites. E.g. is there a statistic going in for the outages of each, don't think so.

    Not worth spending any cent on this. Better ask the vendors for proofs and make your own experience.

  10. ls3901

    About the criteria

    I saw the report has this criteria for high-end array:

    More specifically, high-end arrays must meet the following criteria:

    Single electronics failures are:

    Invisible to the SAN and connected application servers

    Impact less than 25% of the array's performance/throughput

    Microcode updates are:

    Nondisruptive and can be nondisruptively backed out

    Impact less than 25% of the array's performance/throughput

    But DDN SFA12K、HDS HUS VM and IBM DS8870 all are a dual-controller arrays. They do not fit above criteria "Impact less than 25% of the array's performance/throughput" when controller failure.

    Why Gartner select them as the high-end arrays. Anybody tell me why? Thanks.

  11. Lorddraco

    bias ?

    Funny how they include HUS VM and some other array that obviously not meeting the criteria they meet. It is either bias or they do not know that both DDN SFA12K, HUSVM and DS8870 is not dual-controller?

    Another note is are they comparing the OLD EMC VMAX with the rest of the pack while there are VMAX 10K, VMAX 20K (Old VMAX0 and the newer VMAX 40K.

    mmmm ...

    1. Anonymous Coward
      Anonymous Coward

      Re: bias ?

      "Another note is are they comparing the OLD EMC VMAX with the rest of the pack while there are VMAX 10K, VMAX 20K (Old VMAX0 and the newer VMAX 40K."

      Not that old line again......that's not really the case, they're comparing the VMAX family "As of March 2014" and they specifically call out EMC's 16 ports per engine limit, which applies to the newer models you mention above.

      But I do agree that there were a few products that probably shouldn't be in this evaluation.

This topic is closed for new posts.

Biting the hand that feeds IT © 1998–2019