Skip to main content

IBM Netezza Data Warehouse Appliances a big single user box

For those considering to use IBM Netezza Data Warehouse Appliances to set up a data warehouse it might be good to do some testing before stepping into the IBM solution. 

In a single user mode the appliance performs outstanding but as soon as the number of users increase the performance decreases. Considering the fact that you will do a substantial investment in a data warehouse solution for your company that might be a bit of a disappointment to find out that the data warehouse is actually a single user solution. 

Bull Fast Track does perform very well with multiple users firing simultaneously their queries. May be you should take that into consideration when you buy your solution.

Comments

Popular posts from this blog

Privacy and the liberty to express yourself on LinkedIn

Unaware that LinkedIn has such a strong filtering policy that it does not allow me posting a completely innocent post on a Chinese extreme photography website I tried to post the following: "As an Mpx lover I was suprised to find out that the M from Million is now replaced by the B from Billion. This picture is 24 Bpx! Yes you read this well, 24 billion pixels.  Searching on the picture I stumbled on a fellow Nikon lover. If you want to search for him yourself you can find him here: http://www.bigpixel.cn/t/5834170785f26b37002af46a " In my eyes nothing is wrong with this post, but LinkedIn considers it as offending. I changed the lover words, but I could not post it.  Even taking a picture and post it will not let this pass:  Or my critical post on LinkedIn crazy posting policy: it will not pass and I cannot post it.  The technology LinkedIn shows here is an example what to expect in the near future.  Newspapers will have a unified re...

How to run SQL Server 2016 with In-Databasse R on Windows 2016 CTP5

For those who like me tried to run SQL Server 2016 with In-Database R might have run into the same problem as me: In-Database R or the LaunchPad service gives a timeout and won't start. I did several clean installations with different configuration options - for instance I like to put my data on another disk than the system disk - but in the end I tried to do the next, next, next, finish install to see if it something in the setup options is hard coded in there (yes, it happens developers!). For some reason this problem is related to Windows 2016 and not on Windows 2012R2 and I hope the SQL Server team will soon resolve these issues because they are in one word a bit sloppy.  There are 2 issues (maybe even 3 so I give this one also):  The R setup does not create the ExtensibilityLog directory in the "C:\Program Files\Microsoft SQL Server\MSSQL13.MSSQLSERVER\MSSQL\Log" directory The R setup sets the number of users in the SQL Server Launchpad service to 0 it is pos...

Truncate a table in SQL Server and nobody will ever see it....

Today I ran into a very strange problem. I was doing an audit because a client lost 32 billion records in a table and he wanted to know "Who did it!". They did not use auditing because it had a too large impact on performance. The server and instance was not restarted so the DMV's were still available with data.  I presumed there were two options how the data could have been lost and so quickly: run a TRUNCATE TABLE. Very fast and and in less than a second your table is unrecoverable empty  Switch a partition out into a temporary table and just truncate the original table. Yes it happens and the effect is the same: nothing!  So I ran this query to collect the last queries: SELECT t.TEXT QueryName, s.execution_count AS ExecutionCount, s.max_elapsed_time AS MaxElapsedTime, ISNULL(s.total_elapsed_time / s.execution_count, 0) AS AvgElapsedTime, s.creation_time AS LogCreatedOn FROM sys.dm_exec_query_stats s CROSS APPLY sys.dm_exec_sql_text( s.sql_handle ) t WHERE t.TE...