back to article CDOT relatively crap for flash, hyperscalers crap for constant storage

Chief NetApp techie Jay Kidd had some strong words for the flash and cloud crowd at a Wells Fargo event for investors. Among other things, he covered Clustered Data ONTAP and flash and the customer spending slowdown. Not a great fit and a transitory slowdown were the take-away messages. The event was the Wells Fargo Tech …

COMMENTS

This topic is closed for new posts.
  1. Jim O'Reilly

    Those pesky Excel spreadsheets.

    Jay's spreadsheet needs some work on it. S3 is indeed 2.5 cents per gigabyte per month, but that is for 3 replicas of the data, so it's really only 0.8 cents per drive gigabyte. At 3 months that nets 2.4 cents per drive gigabyte, not the 90 cents Jay talks up. This is $32/terabyte/year

    At $32 per year, I suspect that AWS is way lower than the purchase price of a terabyte from NetApp, amortized over 4 years, NetApp has to include the drive, caddy, and a piece of the filer hardware and software to get a fair comparison, and also factor the cost of money.

    1. Anonymous Coward
      Anonymous Coward

      Re: Those pesky Excel spreadsheets.

      Well well Ontap isn't really flash optimized, best let the sales force know.

    2. JohnMartin

      Re: Those pesky Excel spreadsheets.

      -Disclosure NetApp Employee, Opinions are my own -

      The math Jay used is still good, he needed to keep it simple to get the point across. When you do a comparison of $/usable GB over a 3 year lifetime with current pricing AWS still comes off fairly expensive. Even if you assume that S3 uses 3x copies (which AFAIK isn't true, I'm pretty sure that they're using some form of erasure coding now for data resiliency), then that actually skews the game in favour of NetApp who are able to squeeze a lot more effective capacity out of fewer disks with higher levels of data resiliency (5x9s vs 4x9s availability)

      Using list pricing from the 8040 SPC result you'll see about $7K/TiB for high performance storage over a 3 year lifespan, that compares to about $4K/TiB for EBS ... now if you assume the same kinds of $50% "street price" discounts that other vendors use on their SPC submissions, its not hard to be cheaper than amazon.

      These are of course simple models that don't account for Decreasing $/GB costs over the three year period, DC space, power cooling, the relative costs of administration, data protection etc etc, but even more complex models that do account for this still show a competitive advantage for NetApp storage vs S3 and EBS over certain capacity thresholds which are fairly typical for most customers who purchase NetApp's midrange controllers.

    3. Anonymous Coward
      Anonymous Coward

      Re: Those pesky Excel spreadsheets.

      In the way you describe, the math is correct, but one important piece of information is missing. Much of AWS, in particular the private cloud version of AWS which the larger corporations and government agencies who are concerned about data security and will not use a public cloud provider runs on NetApp. So NetApp wins either way whether the customer continues to purchase form them or moves to AWS private cloud.

  2. ByteMe
    Thumb Up

    Nice to see a legacy storage vendor who actually admits that swapping HDDs for SSDs does not make them a flash player. Doesn't take a genius to figure out the same is true for EMC, HP, IBM, etc...

    However, until FlashRay shows-up and actually demonstrates it can do something, NetApp is going to lose market share to flash startups. The EF550 isn't going to win a lot of races (even if they did win against XtremeIO....once....). And don't be fooled by Jay Kidd's comments about not being worried about flash startups because they have big customer accounts. NetApp can't afford to lose *any* accounts right now, and losing lots of small-to-medium customers to flash startups will definitely hurt their bottom line.

    1. Anonymous Coward
      Anonymous Coward

      Genius

      You sure about HP ?

      900k IOps @ <7ms, Increased capacity per SSD, Adaptive sparing for flash, 5 year warranty on MLC, Full field proven data services, a scale out controller architecture as well as all the other flash specific optimizations they made previously.

      http://h30507.www3.hp.com/t5/Around-the-Storage-Block-Blog/HP-3PAR-StoreServ-7450-marching-on-with-all-flash-success/ba-p/158366#.U0RisvldVqI

    2. JohnMartin

      cDOT and EF550 are very much All Flash players.

      -Disclosure NetApp Employee - Opinions are my own -

      1. The All flash array markets is made up of two very different segments, the performance oriented segment (e.g. EF550, 3PAR, Violin, and the old TMS RAMSAN line that IBM bought), and the capacity optimised segment (Pure, Skyera, XtremeIO and All Flash FAS).

      2. The vast majority of flash is being deployed in hybrid arrays, or as DAS inside of servers. The market for all flash arrays is still very smal which partially explains violins implosion. At this point in time, each of the vendors you derided probably sell individually more flash per month than the rest of the "All flash" array players combined. All of them are very much "Flash Players", with bigger annual R&D budgets in flash than many of the startups entire VC funding bankroll.

      3. The EF540 and EF550 are doing nicely, winning pretty much every head to head performance bake-off I'm aware of vs the "performance oriented" all flash offerings out there for sub 1ms consistent price-performance. Put it up against a capacity optimised array with built in dedup/compresson and it simply blows them out of the water. If inline dedup is super-important to you, then combine it with something like Atlantis and you'd be surprised at how good the result is in terms of $/GB and $/IOP, or you can opt for an integrated offering like the All flash FAS.

      4. While an all flash FAS isn't as fast as a performance optimised AFA like the EF540/EF550, its still very fast indeed, certainly fast enough to compete with other capacity optimised all flash arrays, which is partly why it beat out XtremeIO.

      1. Anonymous Coward
        Anonymous Coward

        Re: cDOT and EF550 are very much All Flash players.

        Looks like someone kicked the Netapp ant hill :-)

        FAS - Capacity Optimized ? Shurely shome mishtake....

        1. Anonymous Coward
          Anonymous Coward

          Re: cFAS - Capacity Optimized ? Shurely shome mishtake....

          Tell me another tier 1 storage vendor where you can typically get 100% or more effective usable capacity out of the raw capacity you purchased on both production and disaster recovery/backup. Typical NetApp customers after duplication and compression get back every bit of capacity they give up for RAID parity, hot spares and OnTap overhead. If that is not capacity optimized, I don't know what is.

          1. Anonymous Coward
            Anonymous Coward

            Re: cFAS - Capacity Optimized ? Shurely shome mishtake....

            You've been drinking too much of the marketing kool-aid.

            Please tell me how those arrays perform once both deduplication and compression is enabled across all of those volumes. Oh you don't say ? you don't advise their use across all volumes, only very specific one's where performance isn't such a concern, or where the potential space savings outweigh the performance impact.

            Maybe it's capacity optimized by Netapp's standard but not by the rest of the industry, free capacity does not always equal usable space, this is especially true when it's pinned in the wrong place, Netapp's archaic cluster architecture forces Customers to dedicate resources upfront per controller including, disks, spares, aggregates, volumes and LUN's, never mind all the other WAFL overheads and upfront capacity reserves, which intrinsically pins capacity in the wrong place..

            1. Anonymous Coward
              Anonymous Coward

              Re: cFAS - Capacity Optimized ? Shurely shome mishtake....

              Um, you are seriously misinformed or simply spreading misinformation about NetApp, FAS, WAFL etc. There are literally hundreds of case studies out on NetApp’s website showing customers running deduplication plus compression on mixed workloads without any performance issues and no not on specific use cases, but on mixed workloads as well. Yes NetApp does advise not to run it across all volumes because some things do not deduplicate, video surveillance for example, and some things do not compress, JPG's for example. Why would you run a task against data that it cannot effect as of course it would reduce performance and yield no benefit. That said workloads like VMWare, File Service, Exchange, VDI deduplicate and or compress quite well. My largest customer has 1.5PB of NetApp running Exchange, VMWare and CIFs for file serving and is running that on SATA with FlashCache in front of it. Not only does it perform well, but they are also getting 80% efficiency by combining deduplication AND compression which means for every 100TB of usable capacity they purchased they are getting 180TB effective. Practically every customer I have who is running mixed workloads is seeing 50% + efficiency form deduplication plus compression and not running into performance problems. I keep asking the question of all the NetApp haters. If NetApp sucks so bad why have they been growing every year for the past 8 years and are the #2 storage company in the world?

              1. Anonymous Coward
                Anonymous Coward

                Re: cFAS - Capacity Optimized ? Shurely shome mishtake....

                "Netapp haters"

                It's this quasi religious paranoia that sets Netapp proponents apart and probably also the reason they attract so much stick. :-)

          2. Anonymous Coward
            Anonymous Coward

            Re: cFAS - Capacity Optimized ? Shurely shome mishtake....

            Before making sweeping statements such as this "You can typically get 100% or more effective usable capacity", can you please ensure you provide some independent evidence for such claims, otherwise it's just marketing fluff.

            1. Anonymous Coward
              Anonymous Coward

              Re: cFAS - Capacity Optimized ? Shurely shome mishtake....

              Easy, go to NetApp's reference website, there are literally hundreds of case studies out there stating exactly that. Why would hundreds of NetApp customers put their companies logo on a NetApp case study stating these things if it were not true?

              1. Anonymous Coward
                Anonymous Coward

                Independent !

                I could pull similar case studies for every vendor out there. And Ive seen the overhead of off the shelf compression first hand and you admit to performance overhead above for dedupe even if it is only transient. This is a case of Netapp protesting too much, regarding both flash performance and efficiency, the reality being they're not particularly good at either, remember all these T's & C's around their 50% guarantee.

    3. Anonymous Coward
      Anonymous Coward

      RE: ByteMe

      I think you may have taken Kidd's statement out of context. While the article said he commented "Competition from smaller guys is generally in the smaller accounts", which is true, the article also said " Kidd is not seeing NetApp's growth hobbled by storage startups so there's no need to respond in any special way to them"

      This certainly doesn't mean that NetApp doesn't care about business or competition in smaller accounts. Simply put NetApp is not losing small and medium accounts to these start-ups as NetApp can beat them on features with FAS and on performance and price with E-Series or EF550, so as the article states "there's no need to respond in any special way to them".

  3. Cloud 9

    Nothing to see here ... carry on

    Let's break this down.

    Flash is hugely disruptive and important, delivering extraordinary levels of performance. NetApp admit that CDoT is crap for flash. Their current strategy is to get people to invest in CDoT (or else the last decade of RnD is a complete waste) and operate in a non flash friendly environment for the next 4,5 or 6 years? And what - pretend that flash isn't happening?

    You simply can't charge their prices and be all about the capacity .... because the moment you spend the big bucks and the thing doesn't perform, there are some awkward questions coming from the other side of the finance directors desk (read awkward questions as hairdryer moments).

    This feels like a car crash in slow motion to me.

    I hope you all feel that I adequately broke that down, as per my heartfelt promise at the start.

  4. StorageEngineer

    All-flash ONTAP systems winning against XtremIO

    Is this really true??? Can NetApp show any proof of this? If this true, forget about Pure or XtremeIO or any other pure flash product and NetApp should lead the all flash market...

    Honestly title of this article does not make sense if All flash ONTAP FAS is winning against Xtreme IO. It is not possible that CTO of company is calling their own product as crap....

    1. Anonymous Coward
      Anonymous Coward

      Re: All-flash ONTAP systems winning against XtremIO

      When you combine the total sales of FlashCache, FlashPools, FlashArray, E-Series SSD Cache and EF540/EF550 storage, NetApp sells more Flash than all other storage vendors combined. While you may not consider all or any of these products true flash, they are used in the same manner to provide more performance than spinning disk can alone. NetApp may not always be beating Flash with Flash, but because NetApp has this broad portfolio of choices to address performance challenges, they are beating the competition, not always, but more often than not which is the point Kidd was trying to make.

      1. Anonymous Coward
        Anonymous Coward

        This is just conjecture based on marketing stats and inherrent bias, not facts ! Simply repeating it won't make it come true,.

        1. Anonymous Coward
          Anonymous Coward

          This is just conjecture based on marketing stats

          You are correct, simply repeating it won't make it come true, but IDC showing NetApp has grown storage market share every year for the past 8 years certainly indicated that it is true.

          1. Anonymous Coward
            Anonymous Coward

            Re: This is just conjecture based on marketing stats

            Maybe you need to take off those rose tinted glasses, many of your Colleagues and Customers seem to have already recognized Netapp is not the force it once was.

            "IDC’s quarterly disk storage factory shipments tracker shows NetApp losing ground for three quarters in a row"

            http://www.theregister.co.uk/2014/03/13/idc_storage_tracker_tracks_netapp_tumble/

            "Chris Mellor recently covered IDC’s quarterly disk shipment report, which measures the market leaders in enterprise storage. As Chris points out, NetApp continues to decline, with reduced revenue over the last three quarters"

            http://blog.architecting.it/2014/04/14/has-netapp-reached-a-strategic-inflection-point/

            1. Anonymous Coward
              Anonymous Coward

              Re: This is just conjecture based on marketing stats

              I exchanged some commentary with Chris Mellor directly on that article. Net is three quarters don't make a year and for calendar year 2013 NetApp GAINED share and is actually up 1% compared to calendar year 2012 NetApp sales were down the first three quarters of 2013 mainly due to the US Government spending sequester, not because they lost share to anyone. That is what happens when you sell 42% of the storage to the US Federal government, more than any other storage vendor. Unfortunately without the details of why the first three quarters were down and the full year view, the article makes it sound like NetApp lost share when in fact they still gained despite the temporary shutdown of US Government spending. No rose colored glasses or conjecture based on marketing stats required.

      2. Cloud 9

        Re: All-flash ONTAP systems winning against XtremIO

        "Broad portfolio of choices" .. ughhh I'm already sick of hearing this. How about "large range of plugs to fill the holes"? It's more gritty and down to earth.

        The reason I find it irksome is that a couple of years ago the brag was "one OS fits all" and we were all laughing at the mess of EMC products needed to do what Ontap could do. Now it's a strength and not a weakness.

This topic is closed for new posts.

Other stories you might like