Comments on: SQL SERVER – Identify Most Resource Intensive Queries – SQL in Sixty Seconds #029 – Video

By: Harsh

Harsh — Wed, 17 Oct 2012 05:20:54 +0000

Yes, In modified script, type=2 portion is missing from right join , this change will be enough to get count of non-clustered indexes

By: Vijay Aggrawal

Vijay Aggrawal — Tue, 16 Oct 2012 19:13:46 +0000

The count of nonclustered index is actually returning total count of index type which includes Heap, clustered, non clustered.

SELECT
[schema_name] = s.name
,table_name = o.name
,MAX(i1.type_desc) ClusteredIndexorHeap
,MAX(COALESCE(i2.NCIC,0)) NoOfNonClusteredIndex
,p.rows
FROM sys.indexes i
RIGHT JOIN sys.objects o
ON i.[object_id] = o.[object_id]
INNER JOIN sys.schemas s
ON o.[schema_id] = s.[schema_id]
LEFT JOIN sys.partitions p
ON p.OBJECT_ID = o.OBJECT_ID AND p.index_id IN (0,1)
LEFT JOIN sys.indexes i1
ON i.OBJECT_ID = i1.OBJECT_ID AND i1.TYPE IN (0,1)
LEFT JOIN (
SELECT object_id,COUNT(Index_id) NCIC
FROM sys.indexes
where type = 2
GROUP BY object_id) I2
ON i.OBJECT_ID = i2.OBJECT_ID
WHERE o.TYPE IN (‘U’)
GROUP BY s.name, o.name, p.rows
ORDER BY schema_name, table_name

Now the column of NoOfNonClusteredIndexes will return count of non clustered indexes only.

The query may be further optimized.

By: sandeep

sandeep — Thu, 11 Oct 2012 04:57:30 +0000

Hi sir , can you please help me on below requirement

I had two servers

server 1 ——-SSIS Package is present
server 2—– EXe location

I had SSIS Package in server 1 , at one step am calling the exe present in server 2 and passing arguments and working directory to server 2

when i execute the package the package should execute in Server 1, but when it trigger to exe the exe should execute in server 2 not in server 1

can any one help me on these i can use dotnet also

thanks in advance

By: Rithu

Rithu — Wed, 10 Oct 2012 08:13:57 +0000

I think rather than the toal number of non clustered indexes, it returns the total number of statistics on that table !

By: Rémi BOURGAREL

Rémi BOURGAREL — Wed, 10 Oct 2012 08:02:52 +0000

The video is available anymore, can you send it again ?

By: pinaldave

pinaldave — Wed, 10 Oct 2012 05:57:21 +0000

Fixed. You have a big High Five from me!

By: Harsh

Harsh — Wed, 10 Oct 2012 05:50:57 +0000

One more change required in your script.
Joining Criteria of i1 should have o.OBJECT_ID in place of i.OBJECT_ID

By: pinaldave

pinaldave — Wed, 10 Oct 2012 05:41:10 +0000

Harsh,

I do not know how did I miss that. :-) Fixed!

By: Harsh

Harsh — Wed, 10 Oct 2012 05:26:47 +0000

This script returns only details of tables which do not have any non-clustered indexes.

Script should be:

SELECT [schema_name] = s.name, table_name = o.name,
MAX(i1.type_desc) ClusteredIndexorHeap,
COUNT(i.TYPE) NoOfNonClusteredIndex, p.rows
FROM sys.indexes i
RIGHT JOIN sys.objects o ON i.[object_id] = o.[object_id] AND i.TYPE=2
INNER JOIN sys.schemas s ON o.[schema_id] = s.[schema_id]
LEFT JOIN sys.partitions p ON p.OBJECT_ID = o.OBJECT_ID AND p.index_id IN (0,1)
LEFT JOIN sys.indexes i1 ON o.OBJECT_ID = i1.OBJECT_ID AND i1.TYPE IN (0,1)
WHERE o.TYPE IN (‘U’)
GROUP BY s.name, o.name, p.rows
ORDER BY schema_name, table_name

By: Satya Chaitanya

Satya Chaitanya — Wed, 10 Oct 2012 04:15:38 +0000

Very informative….

what is meant by logical sorting and physical sorting of data in terms of index..