back to article HP’s Mr 3PAR, David Scott, is retiring

Sources close to the situation tell El Reg David Scott, head of HP’s storage business and former boss of 3PAR, is retiring in March. David Scott was CEO of 3PAR when HP bought it for $2.4 billion in September 2010 after a hard-fought bidding war with Dell. It gave HP the storage answer to replace the failing EVA SAN array …

  1. Nate Amsden

    I met david

    For the first time in vegas last year at discover. Really cool guy. I had no idea what to expect. He said he reads all my comments on el reg, so if you're reading this David you rock!!

    Nate

  2. Anonymous Coward
    Anonymous Coward

    Leave before it all crashes down

    HP has no future in storage, neither do the other legacy array companies. Netapp's dying, EMC is re-inventing itself. HDS is incapable of marketing their products. IBM will sell off their tech as soon as they find a buyer.

    Commodity hardware and software stacks will win. It's just a matter of time. Funny, that while Tape still lives on, it will be the spinning disk that hits the grave first.

    1. Anonymous Coward
      Anonymous Coward

      Re: Leave before it all crashes down

      What a load of drivel. Commodity hardware solutions currently don't do even a fraction of the things that real storage solutions can, nor are they able to deliver the same performance without buying almost identical hardware to that of the dedicated storage solution which, by the way, ends up costing pretty much the same in the end.

      All of the marketing around these new solutions is based on magical rainbow unicorn farts and it takes very little knowledge to work out where and how it will fail to deliver.

      Moving to the cloud is a different story, and those guys use commodity hardware for different reasons to the rest of us.

    2. Anonymous Coward
      Anonymous Coward

      Re: Leave before it all crashes down

      Which commodity hardware supports live migrations, storage tier based optimization of live data, thin provisioning, deduplication before remote copying, proactive 24-7 monitoring and replacement of any faulty parts, no questions asked?

      Seriously, I want to know. I would get a big bonus for this.

      1. Anonymous Coward
        Anonymous Coward

        Re: Leave before it all crashes down

        There is no Hardware that supports those items, its the Software that does it, and that can be leveraged in pretty much any well designed distributed storage model. Why bother to even build a technical solution for an organization if its beholden to some specialized hardware construct that locks you into obscene support costs and 35k per TB for a 3 year support model. But hey keep making those "table stakes" legacy array arguments that gets brought up when you want to compete against distributed systems but can't. Oh and "no questions asked" yeah thats legit, after 10 hours on support calls justifying to a 3rd world call center jockey working off a script.

        Legacy array architecture is dead, the only people who still need it are FICON/Mainframe shops. Everyone else will move towards commodity/distributed systems (at least the ones who actually have legitimate IT budgets). Keep a tab on those YoY losses that the big boys keep racking up and tell me that Rome isn't burning.

        1. This post has been deleted by its author

        2. Anonymous Coward
          Anonymous Coward

          Re: Leave before it all crashes down

          Google latency, IOPS, data consistency and throughput and then take another look at your commodity hardware solution. there are very good reasons why some of us use dedicated hardware for storage. I've thoroughly evaluated distributed solutions, including dedicated ones such as left hand, ones running on "commodity" (yet strangely proprietary) hardware such as Nutanix, software defined such as Atlantis and many others, and they all suffer from the same design flaws which make them acceptable for VDI workloads where the data isn't valuable but not for real workloads such as databases where the data is valuable. To make them work for real workloads you need to buy at least as much hardware as a dedicated SAN, usually twice as much to protect data properly since they are effectively single controller until you add a second box. I'm not saying they don't have their place, but that place is certainly not replacing real SAN any time soon for anyone who knows what they are doing.

        3. Lusty

          Re: Leave before it all crashes down

          "Oh and "no questions asked" yeah thats legit, after 10 hours on support calls justifying to a 3rd world call center jockey working off a script."

          Actually all of the proper SAN vendors now replace hardware without you calling them at all. They just turn up at reception with new drives or controllers and you either fit them or let the engineer fit them. If you're calling up a script drone for 10 hours then you're not talking about real SAN hardware!

  3. Anonymous Coward
    Anonymous Coward

    True gent

    I echo the sentiments of a previous comment and the article. He was a true gentleman when I was fortunate enough to attend an exec dinner with David, well rounded (especially considering his bank balance), and cultured. I'm sure he'll be a loss to HP but he has passed on a solid product that has clearly done well since the acquisition. Made a serious amount of dosh, and it couldn't of happened to a nicer guy.

    Just a shame this article has now been reduced to usual FUD commentary!

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2021