A few questions on Table and Index Partitioning in SQL2K5.

Howdy all. I've got a couple questions regarding Partition Function (PF) and
Partition Scheme (PS) and I was wondering if someone could assist?
1. How do I assign a table to a Partition Scheme? I created a test PF and
PS, but how could they really partition my table if the table doesnt know to
use them? I know I can assign the table to a filegroup that has a PS on it,
but didn't know if that alone would cut it?
2. Is there some place I can verify all my settings once I think I have them
right?
3. Would it ever make sense to partition an index without partitioning the
table it resides on?
TIA, ChrisR
I got number 1, but could still use help with 2 and 3. Here is number 1.
CREATE PARTITION FUNCTION myRangePF2 (int)
AS RANGE LEFT FOR VALUES (1, 100, 1000);
GO
CREATE PARTITION SCHEME myRangePS2
AS PARTITION myRangePF2
TO ( fgtest1, fgtest1, fgtest1, fgtest2 );
CREATE TABLE [dbo].[PartitionTest](
[Table1ID] [int] IDENTITY(1,1) NOT NULL,
[TransactionNumber] [int] NOT NULL)
ON myRangePS2 (TransactionNumber)
"ChrisR" <noFudgingWay@.NoEmail.com> wrote in message
news:e7gBSYEBHHA.4212@.TK2MSFTNGP02.phx.gbl...
> Howdy all. I've got a couple questions regarding Partition Function (PF)
and
> Partition Scheme (PS) and I was wondering if someone could assist?
> 1. How do I assign a table to a Partition Scheme? I created a test PF and
> PS, but how could they really partition my table if the table doesnt know
to
> use them? I know I can assign the table to a filegroup that has a PS on
it,
> but didn't know if that alone would cut it?
> 2. Is there some place I can verify all my settings once I think I have
them
> right?
> 3. Would it ever make sense to partition an index without partitioning the
> table it resides on?
> TIA, ChrisR
>
|||> 2. Is there some place I can verify all my settings once I think I have
> them right?
You can test the partition function using the $PARTITION and sample values:
SELECT $PARTITION.myRangePF2(0);
SELECT $PARTITION.myRangePF2(1);
SELECT $PARTITION.myRangePF2(2);

> 3. Would it ever make sense to partition an index without partitioning the
> table it resides on?
Off the top of my head, I guess you might do this if you had a heap (table
with no clustered index) and partitioned only for index manageability. In
most cases you'll have a clustered index and align as well.
Hope this helps.
Dan Guzman
SQL Server MVP
"ChrisR" <noFudgingWay@.NoEmail.com> wrote in message
news:eqSA72EBHHA.4864@.TK2MSFTNGP04.phx.gbl...
>I got number 1, but could still use help with 2 and 3. Here is number 1.
> CREATE PARTITION FUNCTION myRangePF2 (int)
> AS RANGE LEFT FOR VALUES (1, 100, 1000);
> GO
>
> CREATE PARTITION SCHEME myRangePS2
> AS PARTITION myRangePF2
> TO ( fgtest1, fgtest1, fgtest1, fgtest2 );
>
> CREATE TABLE [dbo].[PartitionTest](
> [Table1ID] [int] IDENTITY(1,1) NOT NULL,
> [TransactionNumber] [int] NOT NULL)
> ON myRangePS2 (TransactionNumber)
>
>
>
> "ChrisR" <noFudgingWay@.NoEmail.com> wrote in message
> news:e7gBSYEBHHA.4212@.TK2MSFTNGP02.phx.gbl...
> and
> to
> it,
> them
>
|||Thanks Dan. To piggyback on your idea, I also just stumbled on the
sys.partition_* functions that are proving to be useful.
"Dan Guzman" <guzmanda@.nospam-online.sbcglobal.net> wrote in message
news:56080A38-716B-4144-8694-616FA7B84903@.microsoft.com...
> You can test the partition function using the $PARTITION and sample
values:[vbcol=seagreen]
> SELECT $PARTITION.myRangePF2(0);
> SELECT $PARTITION.myRangePF2(1);
> SELECT $PARTITION.myRangePF2(2);
the[vbcol=seagreen]
> Off the top of my head, I guess you might do this if you had a heap (table
> with no clustered index) and partitioned only for index manageability. In
> most cases you'll have a clustered index and align as well.
> --
> Hope this helps.
> Dan Guzman
> SQL Server MVP
> "ChrisR" <noFudgingWay@.NoEmail.com> wrote in message
> news:eqSA72EBHHA.4864@.TK2MSFTNGP04.phx.gbl...
(PF)[vbcol=seagreen]
and[vbcol=seagreen]
know
>