finding minimum value

How to find the minimum value from time Timestamp column ?

I want get the eralier timestamp vlaues from the avaliable list

when iam using Aggregate transformation ..its again giving all the list of vlaues

Thanks
Niru

If you have only one minimum per data set I would use the Script component to get it manually.

Thanks.

|||no i have lot of set of duplicate values with in the same column....I have to get the Minimum ones in those sets ..like

5
5
3
3

Desired result:

5
5

|||

I still do not understand your scenario.

If you have a data like this:

A B

1 1

1 2

2 7

2 8

You can use the Aggregate component to group on column A and find a minimum on B, so it would produce:

A B

1 1

2 7

Is that what you are looking for?

Finding displaying and deleteing dulpicate records

HI all,
I am trying to find duplicate records ina database that I have taken over. I
am using group bys and count() to find the sups.
I also realise that we could have a problem with misspellings and hterefore
not have a duplicate as in Joe Soape, Jo Soap, Joe Sope Joe Soap. These
couldpossible be sups of the same. I would terefore assume to allow for user
to decide what is dups and what to keep.
Are there any other more acceptable ways and means of doing this.
THis a a sample of what I use
SELECT Name, address1, address2,city, COUNT(*) as numbofdups FROM reinsurer
GROUP BY Name, address1, address2, city HAVING COUNT(*) > 1
Thanks
RobertIf you are using SQL 2005 developer / enterprise edition you could use the
Fuzzy Grouping / Term Extraction transformations in Integration Services and
work it out through that.
Aside from that you need to role your own in T-SQL, there are functions like
SOUNDEX that help to an extent but aren't that brilliant.
Tony.
Tony Rogerson
SQL Server MVP
http://sqlserverfaq.com - free video tutorials
"Robert Bravery" <me@.u.com> wrote in message
news:u0uuiXbbGHA.3352@.TK2MSFTNGP03.phx.gbl...
> HI all,
> I am trying to find duplicate records ina database that I have taken over.
> I
> am using group bys and count() to find the sups.
> I also realise that we could have a problem with misspellings and
> hterefore
> not have a duplicate as in Joe Soape, Jo Soap, Joe Sope Joe Soap. These
> couldpossible be sups of the same. I would terefore assume to allow for
> user
> to decide what is dups and what to keep.
> Are there any other more acceptable ways and means of doing this.
> THis a a sample of what I use
> SELECT Name, address1, address2,city, COUNT(*) as numbofdups FROM
> reinsurer
> GROUP BY Name, address1, address2, city HAVING COUNT(*) > 1
> Thanks
> Robert
>
>