July 30, 2009

“The Netezza price point”

Over the past couple of years, quite a few data warehouse appliance or DBMS vendors have talked to me directly in terms of “Netezza’s price point,” or some similar phrase. Some have indicated that they’re right around the Netezza price point, but think their products are superior to Netezza’s. Others have stressed the large gap between their price and Netezza’s. But one way or the other, “Netezza’s price” has been an industry metric.

One reason everybody talks about the “Netezza (list) price” is that it hasn’t been changing much, seemingly staying stable at $50-60K/terabyte for a long time. And thus Teradata’s 2550 and Oracle’s larger-disk Exadata configuration — both priced more or less in the same range — have clearly been price-competitive with Netezza since their respective introductions.

That just changed. Netezza is cutting its pricing to the $20K/terabyte range imminently, with further cuts to come. So where does that leave competitors?

Comments

14 Responses to ““The Netezza price point””

  1. Netezza is changing its hardware architecture and slashing prices accordingly | DBMS2 -- DataBase Management System Services on July 30th, 2009 10:14 pm

    […] “The Netezza Price Point” just changed. […]

  2. Jerome Pineau on July 31st, 2009 11:35 am

    What’s always confusing to me is what each vendor means by “per terabyte” – is that terabytes of uncompressed user data? Before or after indexing and all the usual massaging? What if you copy the data on multiple nodes? How do they know how many terabytes you’re loading anyway? Who keeps track?

    Also is Oracle the only one left pricing per core?

  3. Curt Monash on July 31st, 2009 11:51 am

    I’ve posted on that twice in the past two months — e.g. http://www.dbms2.com/2009/07/02/daniel-abadi-user-data/ — and people are still confused …

    Yes, vendors also publish raw disk metrics in many cases — but then they don’t call it “user data”.

    Also http://www.dbms2.com/2008/09/01/estimating-user-data-vs-spinning-disk/

  4. Paul Johnson on July 31st, 2009 4:35 pm

    The cost of the host DBMS and server is pretty insignificant in the Dataupia setup. A single CPU machine running Oracle or SQL Server standard edition is all that’s needed.

    MPP vendors don’t charge per core as they use too many of them.

    The newer entrants into the market have led the recent move to per TB pricing.

    The Teradata products with competitive pricing are not balanced sytems, so the comparisons are invalid imho.

  5. RC on August 1st, 2009 4:29 pm

    @Curt Monash

    Analytic databases are nowadays either a row store or a column store.

    Do you think that in the future there will be analyic db’s that you can use both as a column store and as a row store? Or that you can even decide to store some attributes of a table in classic row-format and other in a column-format?

  6. Curt Monash on August 4th, 2009 12:03 pm
  7. RC on August 4th, 2009 1:42 pm

    I noticed your new article. There is and was something in the air. Maybe Oracle and MS SQL will provide hybrid solutions too?

  8. Curt Monash on August 4th, 2009 1:47 pm

    At a company as large as Oracle, Microsoft, or IBM, for any sufficiently reasonable strategy, there exists at least one fairly senior employee who not only believes it SHOULD be followed but indeed believes it WILL be followed.

    So while it’s certain they’re looking at the idea, it’s impossible to predict whether they’ll follow through on it, unless one has some kind of NDA-level insight.

  9. Brian Ganly on August 6th, 2009 5:32 am

    User data is a key price for me but user data is a confusing term. How much is actually business data.
    Assuming you could by a 6TB Exadata Machine and a 6TB Netezza server at the old $60k per TB price point and you fill the servers to the very top I calculate that Oracle costs $297k per TB of business data and Netezza costs $60k per TB of business data

    Prices assume Oracle 2xCompression and 50% discount from list, no Compression on Netezza.

    Here how it is calculated.

    System Size 6TB
    Oracle With 2x Compression, Performance Server 300GB disks
    2TB System Data – Temp/Redo/Archive Redo/Undo etc
    2TB “User Data” – Indexes, Aggregates, Empty space in blocks etc.
    2TB Business Data – Data in tables
    (These figures are reflected in our Oracle 9.2 DWH DB we do not have Exadata)

    Netezza No Compression
    6TB Business Data
    Swap space/Mirrors not presented as part of the price. No indexes or aggregates due to server speed.

    Oracle Exa TB $per TB 6TB Cost
    Overall 6 $99 $594
    Business Data 2 $297 $594
    Netezza
    Overall 6 $60 $360
    Business Data 6 $60 $360

    NB Oracle prices calculated according to this link and I gave myself 50% discount.
    http://www.dbms2.com/2008/09/30/oracle-database-machine-exadata-pricing-part-2/

  10. Oracle Exadata 2 capacity pricing | DBMS2 -- DataBase Management System Services on October 5th, 2009 8:22 am

    […] although Netezza recently went with a not-sandbagged 2.25X compression estimate to get below the $20K/terabyte price point. Columnar software vendors have tended to be more aggressive, with figures of 10X or more casually […]

  11. A framework for thinking about data warehouse growth | DBMS2 -- DataBase Management System Services on December 7th, 2009 9:51 am

    […] that “disk is the new tape.” No-apologies performance can be had on data warehouse systems for $20,000/terabyte or less – perhaps even a lot less. Tolerable performance may cost 3-4X less than that. I think a […]

  12. Notes on data warehouse appliance prices | DBMS 2 : DataBase Management System Services on October 15th, 2010 2:53 am

    […] wrote last year that Netezza provides the traditional industry benchmark for per-terabyte pricing. When I wrote that, the “Netezza price point” had just become a little under […]

  13. John on June 4th, 2011 1:56 am
  14. Amazon Redshift and its implications | DBMS 2 : DataBase Management System Services on December 9th, 2012 11:59 am

    […] 3 years ago, $20,000/terabyte was a great list price for purchasing a data warehouse appliance that required little […]

Leave a Reply




Feed: DBMS (database management system), DW (data warehousing), BI (business intelligence), and analytics technology Subscribe to the Monash Research feed via RSS or email:

Login

Search our blogs and white papers

Monash Research blogs

User consulting

Building a short list? Refining your strategic plan? We can help.

Vendor advisory

We tell vendors what's happening -- and, more important, what they should do about it.

Monash Research highlights

Learn about white papers, webcasts, and blog highlights, by RSS or email.