-1

We are busy with an project for school. we have to build an Datawarehouse but im stuck.

I have the following starschema:

dim_location:

Location_key(Primary key)

City

dim_condition:

Condition_key(primary key)

Description(Example: Good, Bad, Very Bad, Very Good etc.)

dim_time:

Time_key

Year

Date

Quarter

etc.

Fact table Ground:

Time_Key(foreign)

Location_Key(foreign)

Condition_Key(foreign)

The location dimension should be an table with al kinds of locations. The condition dimension should be an table with conditions like good, bad, not so bad, very good etc. And the time dimension is should be an table with all days, quarters etc.

What i should know out of this is the following: "I want to know on a certain date on a certain location the ground had a certain condition"

example: On 20 January on Monday the ground in Amsterdam was not good.

I have some dumps that i should use to fill those dimension and facts. But i dont know what the best way is, and how to fill the fact table. I have already filled the dimensions in SSIS, That went good but I cant fill the fact table.(Tried: Multiple Lookups, Merge, Merge join) Can someone tell me in steps how i should make the datawarehouse and fill it(The best way)?

Sorry for my bad English.

Help is very appreciated.

Thanx!!

Nordip
  • 29
  • 3

1 Answers1

0

The simplest way to start out is full refresh - that way your design can quickly adapt. So start with an Execute SQL task to truncate your target table. Then add a Data Flow task to reload your Fact table. Start with a Source (typically OLE DB). Add Lookups (typically Full Cache) to get your Dimension Keys, one per Dimension. Finish with a Destination (typically OLE DB, with Fast Load and Table Lock) to insert the data.

As you package matures you can add features like incremental extract and disable/rebuild of indexes.

Good luck!

Mike Honey
  • 14,523
  • 1
  • 24
  • 40