Thanks for the input. Based on a lot of input here I did decided to use bigint for the tables that will get really big.

Aaron West (6/11/2008)

---

And as for...

Thats not a bad idea! My approach would net the same results, but your way would keep it simpler. I think I might just plan on doing that....

Don't worry, I'm not going to be doing any 'like' statements on a vchar field. All the queries will be searching on a User_ID field. (each row will...

Jeff Moden (6/10/2008)

---

I agree with Paul's assessment... it's a good thing to plan on that type of scalability, always... but I see no mention of "clustered" servers or the other...

Paul Lach (6/10/2008)

---

As a database designer/developer one of the most difficult things to do is to differentiate between what the customer/manager SAYS you need to do and what the product/business...

Yes it would. But, from what I've read, having that many records in one unpartitioned table isn't a good thing for a...

Yes it would. However, there will be over 2 billion rows...

That is a very true statement. My plan...

Oh, that makes more sense! Thanks for clearing that one up!

Good point. Unfortunately, its hard to know the answer to that question now. I just wonder how hard it would be for me to determine what I thought was...

Great. Thanks for looking! 🙂

Sorry, maybe I didn't explain that well. What I...

So, I did more research and also a little bit of testing. Here is a working query that does what it is suppose to. Does it look too...

Unfortunately that isn't possible in our system. ...

I couldn't agree with you more!

But we are a small startup company and there just isn't any $$ for a DBA. 🙁