|
Home > Archive > MS SQL Server > January 2006 > optimizing load performance using partitioned tables in 2005
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
optimizing load performance using partitioned tables in 2005
|
|
|
|
Hi,
Iam curious to know if i can increase my load performance using
partitioned tables. If i create a 4-way partitioned table, can i load
directly into a specific partition, so effectively having 4 parallel loads
into the 4 partitions. (Something i can do in Sybase).
If not what can i do to maximize my load on a partitioned table?
Vivek
| |
| Hilary Cotter 2006-01-19, 7:23 am |
| This is one option, but does your data align itself with these partitions?
Supposed you partition on last name, are your last names going to be evenly
distributed? Last names are a pretty good choice as the distribution is
somewhat even, however a choice like date is probably not so good if your
data is ordered already.
--
Hilary Cotter
Looking for a SQL Server replication book?
http://www.nwsu.com/0974973602.html
Looking for a FAQ on Indexing Services/SQL FTS
http://www.indexserverfaq.com
"Vivek" <Vivek@discussions.microsoft.com> wrote in message
news:6196820D-E5C6-4D83-97F1- 0D7087B6AEC6@microso
ft.com...
>
> Hi,
>
> Iam curious to know if i can increase my load performance using
> partitioned tables. If i create a 4-way partitioned table, can i load
> directly into a specific partition, so effectively having 4 parallel loads
> into the 4 partitions. (Something i can do in Sybase).
> If not what can i do to maximize my load on a partitioned table?
>
> Vivek
| |
|
|
Yes my data does align itself with these partitions. So what i want to know
is how do I load data into a specific partition? (using say BCP or Bulk
Insert) What is the syntax?
"Hilary Cotter" wrote:
> This is one option, but does your data align itself with these partitions?
> Supposed you partition on last name, are your last names going to be evenly
> distributed? Last names are a pretty good choice as the distribution is
> somewhat even, however a choice like date is probably not so good if your
> data is ordered already.
>
> --
> Hilary Cotter
> Looking for a SQL Server replication book?
> http://www.nwsu.com/0974973602.html
>
> Looking for a FAQ on Indexing Services/SQL FTS
> http://www.indexserverfaq.com
>
> "Vivek" <Vivek@discussions.microsoft.com> wrote in message
> news:6196820D-E5C6-4D83-97F1- 0D7087B6AEC6@microso
ft.com...
>
>
>
|
|
|
|
|