back to article BI benchmark outs HP Superdome 2 details

More details of HP's forthcoming Tukwila Itanium 9300 servers have come to light thanks to TPC-H data warehousing benchmark test results that HP has just published. When HP announced the first of its Itanium 9300 servers three weeks ago, the company's techies gave out some of the salient characteristics of the new blade- …

COMMENTS

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

    Run that one by me again...

    Sometimes I wonder why some of the quantitative comparison info in El Reg articles isn't presented in a more appropriate format. I mean, I've heard this rumour that you don't need Excel to do tables, that techy geeks can actually do them natively in HTML using Visual Notepad, and that there are other less geeky but equally effective methods of achieving the same result.

    Anyway, after wading through far more text than is actually necessary, the salient points which I'm awaiting the market's reaction on appear to be:

    New Improved Superdome: throughput 140kQPH @ ~$20 per QPH

    Single Proliant (last year's model): throughput 103kQPH @~ $4 per QPH

    Clustered Proliant: throughput 1320kQPH @~$1 per QPH

    Not looking like an ideal Superdome application based on these numbers.

    1. Anonymous Coward
      Megaphone

      x86 scale-out cheaper than UNIX scale up shocker!

      Wow - what amazing insight - are you telling me that some x86 boxes scaled out works out cheaper?? I'd never have thought...

      Seriously, the point of these sorts of systems isn''t to be the cheapest is it?

      1. Anonymous Coward
        FAIL

        "the point of these sorts of systems isn''t to be the cheapest is it?"

        If they're not the cheapest (best value for money) for *something*, what exactly is their point, please? {please don't say enterprise-class RAS unless you can back it up with evidence of stuff that isn't available in Proliant-class boxes]

      2. Jesper Frimann

        Ehh...

        Actually they, or at least some of them are pretty close i price/performance

        http://www.tpc.org/results/individual_results/IBM/IBM_780_TPCC_20100412_es.pdf

        http://www.tpc.org/results/individual_results/HP/hp_DL380_TPCC_051110_ES.pdf

        Comparison on price/performance lowest on this page:

        http://www.ideasinternational.com/benchmark/ben020.aspx?b=eb4a0fa9-0344-487d-85ef-49539f0da8f0&f=Clust%27d%3dN

        Now that is pretty close.

        // Jesper

  2. Anonymous Coward
    Stop

    TPC-H and SAP performance

    HP continues to do 1TB benchmarks which are trivial compared to what most customers want/need.

    Here is the SAP performance data as reported earlier:

    Tukwila is only four core and the SAP performance measure of each core is only 1,200 SAPs.

    Compare this to 2,333 SAPs for Nehalem and the Power7 SAPs of 3,159

    No wonder there is a 2 for 1 sale.

    VMotion and Power Mobility have been around for 3+ years, and HP wonders why people think they are divesting in the technology.

    Liberty

  3. Anonymous Coward
    Boffin

    almost no IBM Power TPC-H benchmarks

    "IBM has not yet run the TPC-H test using the 1TB data-set size (or any other) on its new Power7-based machines. IBM and HP tend to stay away from each other in the TPC-H test, which has data warehouses that come in 100GB, 300GB, 1TB, 3TB, 10TB, and 30TB sizes. It seems likely that the two will avoid going head-to-head in future comparisons."

    hmmm so they are "avoiding head-to-head comparisons" are they? Or are you drinking IBM kool-aid TPM? Seems more like IBM has pretty much always avoided TPC-H benchmarks for Power - there are a couple out there, but little that lets you really compare... maybe its much harder to game a benchmark where you can't partition so much you fit all your DB operations into single cache-lines (a-la TPC-C)?

  4. Anonymous Coward
    Thumb Down

    Big SMP for TPC-H...

    ...how very last decade!

This topic is closed for new posts.

Other stories you might like