Posts Tagged ‘DW Appliance’

Oracle Enters The DW Appliance Market

Thursday, September 25th, 2008

Well now – has the inevitable happened? Larry Ellison, Oracle’s CEO finally recognised the value of the hardware/software combination in the business analytics marketplace. This market, pioneered over 20 years ago by Teradata and now thriving with many other players including Netezza, ParAccel, Datupia, ExaSol, Vertica and others, has now become a target for Oracle who have clearly had enough of competitors eating away at the Oracle database with lower TCO DBMS offerings optimised for analysis and reporting. With the recent acquisition of DatAllegro by Microsoft, IBM with its Balanced Warehouse and now Oracle entering the database machine market it certainly seems that the DW Appliance market is now becoming a hot competitive battleground.

The newly announced Oracle Exadata DW Appliance is jointly developed by Oracle and HP and will be sold directly by Oracle. The Exadata server runs the Oracle parallel server on Oracle Enterprise Linux. It has 8-HP Proliant DL360 G5 database servers, with

  • 2 quad-core Intel Xeon Processor E5430 (2.66GHz)
  • 32GB memory
  • 1-HP InfiniBand Dual Port HCA
  • 4-146GB SAS 10K hard disk drives
  • 4-24-port InfiniBand switches
  • 14-HP Exadata Storage Server Hardware–each is an HP ProLiant DL180 G5, with 2 quad-core Intel Xeon Processor E5430 (2.66GHz)
  • 8GB memory
  • 1-HP InfiniBand Dual Port HCA
  • 12-300GB SAS or 12-1TB SATA disk drives

My question on this announcement is given that HP are jointly in on the Exadata product offering with Oracle, what does this mean for HP’s own DW Appliance offering – the HP NeoView Appliance? This is also a parallel DBMS product that competes with Oracle. I assume that with HP playing in both markets (its own DBMS product on its own hardware plus the hardware behind the Oracle Exadata offering) that it is seeking to maximising the revenue it can take by covering all bases. Time will tell. In my opinion it is clear that with so many vendors now in the DW Appliance market it is going to take a lot more than just TPC-H benchmarks to get a differential. It certainly means customers will have to look closely at performance claims. Everyone will claim they are the fastest which could easily result in prospective customers demanding more to distinguish one vendor from another. For this reason I believe that analytic application appliances have to happen (analytic application pre-installed on a DW Appliance). Vendors who go deep on vertical analytic application appliances could carve out a very lucrative business when you combine this with the attraction of low TCO DW Appliance offerings.

DatAllegro – The New Microsoft DW Appliance?

Monday, July 28th, 2008

Why is it every time I take a week off on holiday something major happens in the BI market? I am of course talking about Microsoft’s Announcement to acquire the data warehouse appliance vendor DatAllegro. The message on this is certainly obvious, the scaling of SQL Server. It’s a very interesting announcement. With the exception of Netezza who have done well here in the UK and in Europe, many of the DW appliance vendors have been struggling. Almost all of them have been chasing business in the same vertical industries that have high volumes of data (e.g. Telco, Retail, Financial Services). Now the one giant that people were wondering about in terms of parallel DBMS scalability, has moved. However we are clearly going to have to wait to see how well SQL Server scales in this kind of setup. Even prior to this announcement the myth that SQL Server would not scale beyond 1 Terabyte has long been proved incorrect however. I have certainly had clients running single instance SQL Server BI system databases at around 13-15 terabytes for a number of years now. No doubt there are larger configurations than that out there. However this announcement will certainly lift Microsoft customer confidence that Redmond are serious about offering a scalable SQL Server option on non-proprietary hardware that starts to compete with the parallel DBMS offerings of IBM, Oracle and Teradata as well as other DW Appliance offerings. Time will tell what will happen and how competitive this will be. SQL Server Integration Services (the Microsoft ETL tool that ships with SQL Server) will also have to scale however to get larger data volumes on to a parallel SQL Server. So we have to see what Microsoft will do here. This announcement also offers up an interesting option for Kalido who recently announced support to generate for Microsoft SQL Server.

DW Appliance Market Adds a New Player

Tuesday, November 13th, 2007

At TDWI in the US recently, ParAccel became the latest vendor to announce its arrival into the Data Warehouse Appliance market. This market already has players like DatAllegro, Dataupia, GreenPlum , HP , Netezza, and Teradata to name a few.

ParAccel’s new Massively Parallel Analytic Database runs on commonity hardware (first available release is Sun but other releases will follow on Dell, HP and IBM hardware) and comes in two flavours. These are:


ParAccel MAVERICK – a stand alone analytic platform using ANSI SQL
ParAccel AMIGO – a drop-in acceleration platform with query routing, synch, and syntax coverage for existing Microsoft SQL Server and Oracle instances


Interestingly this product works on columns during its parallel query processing rather than rows which raises some observations. For a start, in many of the DW reviews that I have conducted over the years I have seen the well known practice of “Mini Dimensions” introduced to isolate popular columns in a large dimension (e.g. Customer) into their own separate Mini-dimension table. This is a performance ‘trick’ that can often speed up joins between large dimensions and large fact tables whereby if popular columns in the large dimension are selected in a query to qualify metrics in the fact table, then the join occurs between the mini dimension and the fact table instead of the much larger ‘real’ dimension and the fact table. As a result, join processing is faster.

 

The only problem with mini dimensions is that they complicate the key structures of fact tables, all BI tool business views (E.g. SAS Information Maps, MS Report Models, Business Objects Universes etc.) need to know about them in order to generate the right join SQL and tracking history across changes to columns in the main dimension and mini dimensions can be complex. The column approach to parallel query processing taken by ParAccel may well negate the need for mini dimensions as a performance tuning mechanism in many star schemas thereby simplifying design. If you are in the process of selecting a DW Appliance product it would certainly be worth investigating this point and worth a benchmark test.


In addition SQL advanced analytic aggregate functions are all column based. Once again therefore it would be worth benchmarking this in ParAccel Vs other DW Appliance products as there may well be a boost in performance here too for parallel processing at the column level. I might add that I have NOT personally benchmarked this but simply looking at the way this product is architected it would certainly warrant a look for specific analytic applications. Benchmarking is recommended however.

 

The ParAccel Analytic Database software is available now. ParAccel offers two licensing options


$1,000 per gigabyte for all-in-memory systems beginning at 100GB,
or $40,000 per node plus $10,000 per terabyte for disk-based systems beginning at 5 nodes.
Subscription licensing is also available starting as low as $5,000 per month.