data-warehouse

We are currently using a summary table that aggregates information for our users on an hourly basis in UTC time. The problem we are having is that this table is becoming too large and slowing our system down immensely. We have done all the tuning techniques recommended for PostgreSQL and we are still experiencing slowness. Our idea was to start aggregating by day rather than by hour, but the problem is that we allow our customers to change the timezone, which recalculates the data for that day. Does anyone know of a way to store the daily summary but still respect the numbers and totals when

I am writing a query to fetch records from a Oracle warehouse. Its a simple Select Query with joins on few tables and i have few columns to be aggregated. Hence i end up using Groupby on rest of the columns. Say I am picking some 10 columns and out of which 5 is aggregate columns. so i need to group by on the other 5 columns. I can even achieve the same by not doing a Groupby and using over (paritition by) clause on the each each aggregate column i want to derive. I am not sure which is better against a warehouse or in general. They are not the same. This will return 3 rows: select deptno,

We're creating a historic archive for a world history database and we need a date lookup table which references all dates in AD. How to go about creating the values for this table - from 1AD to 2011 as YYYY/MM/DD? Database is MySQL. Problems: I'm using Excel to pre-populate the dates, then import into MySQL as: YYYY/MM/DD but Excel doesn't recognize years like 0007, 0008, etc so I can't auto-copy cells to generate dates. I have to manually do it and this will take days to go from 1AD to year 2011 as YYYY/MM/DD. Leap years were introduced on 1752. If I programmatically generates dates how do I

Where I can download sample database which can be used for data warehouse creation? It should't be sample from Microsoft (Northwind etc.). EDIT: Sorry for not clarifying my question. At my university we have class where we must create some data warehouse and since Northwind is so popular over net then professor told us not to use this database. We will use for this SQL Server 2008 but using Northwind is forbidden. Whatever happened to NOT Northwind? http://www.hanselman.com/blog/CommunityCallToActionNOTNorthwind.aspx There's also SQL Data Generator from Redgate: http://www.red-gate.com

I have heard a few references that pk is not required on fact table. I believe every single table should have a pk. How could a person understand a row in a fact table if there is no pk and 10+ foreign keys. Primary Key is there ... but Enforcing the primary key constraint in database level is not required. If you think about this, technically a unique key or primary key is a key that uniquely defines the characteristics of each row. And it can be composed of more than one attributes of that entity. Now in the case of a Fact table, foreign keys flowing-in from the other dimension tables