Database

MSOLAP.2 provider problem

6-Jun-09 19:01

Yes I fully agree with Scott M

Use distinct clause
Smile | :)

Niladri Biswas

ThaScorpion4-Jun-09 3:26

ThaScorpion

4-Jun-09 3:26

Team,

I have a web Windows server 2003 64-bit and installed Sql server analysis services 2000 along with SP4, the problem is that: our application is ASP classic and has the connection string:
"Provider=MSOLAP.2;Timeout=60;Data Source=localhost;Initial Catalog=myCube;Client Cache Size=25"
I can not find a provider for MSOLAP.2, I have copied most of the dlls from the legacy server (32-bit) and registered them, but still cannot connect to that cube.
when I check with a .udl file I don't see any provider for OLAP.

What do I need to setup? Any idea? What is wrong here?

Note: I can open analysis services manager and connect to my cubes successfully.

please advice.

Thanks in advance.

ThaScorpion

how to handle no data found error in sql server

lakshmichawala4-Jun-09 3:25

lakshmichawala

4-Jun-09 3:25

how to handle no data found error in sql server

Re: how to handle no data found error in sql server

Ashfield4-Jun-09 21:04

Re: One more query problem

4-Jun-09 21:04

How is it an error? It just means youur query returned no results. Handle it where? Please explain exactly what the problem is, my mystic powers are worn out by Friday.

Bob
Ashfield Consultants Ltd
Proud to be a 2009 Code Project MVP

One more query problem

sujithkumarsl3-Jun-09 23:36

sujithkumarsl

3-Jun-09 23:36

Suppose i have a table like this..

COL1    COL2    COL3
X       100	  1
X       200	  2
X       300	  3
Y       100	  1
Y       200	  2
Z       300	  1

i want to select col1 where 100=1 and 200=2 and 300=3

My small attempt...

Mycroft Holmes4-Jun-09 19:42

Mycroft Holmes

4-Jun-09 19:42

In the where clause convert both col2 & 3 to strings and compare the first character of each.

Never underestimate the power of human stupidity
RAH

Re: One more query problem

Niladri_Biswas6-Jun-09 19:38

Re: Query problem [modified]

6-Jun-09 19:38

Solution 1:

SELECT COL1
FROM   tbl_Test
WHERE  COL2 = (COL3 * 100);

Solution 2:

SELECT COL1
FROM  tbl_Test
WHERE (COL2 = 100 AND COL3 = 1)
OR (COL2 = 200 AND COL3 = 2)
OR (COL2 = 300 AND COL3 = 3)

Solution 3:

select COL1 
from tbl_Test
where ( (COL2=100 and COL3=1) or (COL2=200 and COL3=2) or (COL2=300 and COL3=3) )

Please vote Smile | :)

Niladri Biswas

Query problem [modified]

sujithkumarsl3-Jun-09 22:31

sujithkumarsl

3-Jun-09 22:31

i am using sql server database

please check the below table two columns say column A and Column B

if the user give 100,200 and 300 i want return 1 ( which has all these values)
if the user give 100 and 200 i have to give 1 and 2. Hope this is clear.... Please help me to build the query

My small attempt...

modified on Thursday, June 4, 2009 7:50 AM

Niladri_Biswas6-Jun-09 20:01

6-Jun-09 20:01

Hello this time the question is not clear.

Please be more specific.

This is a very poor way of asking your question. Please define the function mapping your input to output more specifically

Tell us the desired output.

Is it

For case 1:

Input 100,200,300

Output:

A  B

1  100
1  200
1  300

For Case 2:
I/P: 100 , 200

?

Please from next time whenever you post, give the sample output so that it will be easy to solve.

Also, are you passing a comma delimited value or what?

Nothing clear!: Frown | :(

Niladri Biswas

modified on Sunday, June 7, 2009 2:39 AM

Re: Query problem

Niladri_Biswas6-Jun-09 20:44

Perfomance on the Following Query

6-Jun-09 20:44

Anyway

Assuming that

For case 1:

Input 100,200,300 

Output:

A  B

1  100
1  200
1  300

Solution

SELECT MIN(A), B
FROM tbl_test
WHERE B IN (100, 200, 300)
GROUP BY B;

For Case 2:

I/P: 100 , 200

O/P:

A  B
1  100
1  200
2  100
2  200

Solution is

SELECT A, B
FROM tbl_test
WHERE B IN (100, 200)
ORDER BY A, B;

hope this helps.

And vote me Smile | :)

And from next time please define the function mapping your input to output more specifically

Smile | :)

Niladri Biswas

Default Collation

hadad3-Jun-09 22:03

hadad

3-Jun-09 22:03

Hello,
When I try to insert unicode chars (Arabic chars) in my database which has its default collation to (latin) these char are inserted incorectly even if I use the nvarchar data type is this because of the collation, can II use the Arabic bin collation to resolve this proplem or ther is an other raison?

Thanks

Dad

Vimalsoft(Pty) Ltd3-Jun-09 20:54

Vimalsoft(Pty) Ltd

3-Jun-09 20:54

Good Morning All

I have Table Defined as

/****** Object:  Table [dbo].[EXP_REL_SLOT_DOMN]    Script Date: 06/04/2009 08:38:19 ******/
SET ANSI_NULLS ON
GO
SET QUOTED_IDENTIFIER ON
GO
CREATE TABLE [dbo].[EXP_REL_SLOT_DOMN](
	[SLOT] [int] NOT NULL,
	[DOMN] [int] NOT NULL,
	[PREF] [int] NOT NULL
) ON [PRIMARY]

GO
EXEC sys.sp_addextendedproperty @name=N'MS_Description', @value=N'references TBL_SLOT_ALLC.ID' ,@level0type=N'SCHEMA', @level0name=N'dbo', @level1type=N'TABLE', @level1name=N'EXP_REL_SLOT_DOMN', @level2type=N'COLUMN', @level2name=N'SLOT'

indexed as

/****** Object:  Index [EXP_REL_SLOT_DOMN_INDEX]    Script Date: 06/04/2009 08:41:01 ******/
CREATE UNIQUE CLUSTERED INDEX [EXP_REL_SLOT_DOMN_INDEX] ON [dbo].[EXP_REL_SLOT_DOMN] 
(
	[SLOT] ASC,
	[DOMN] ASC,
	[PREF] ASC
)WITH (SORT_IN_TEMPDB = OFF, DROP_EXISTING = OFF, IGNORE_DUP_KEY = ON, ONLINE = OFF) ON [PRIMARY]

And i have Two insert statements that takes 29 seconds each

--29 seconds
INSERT into tempslot
select distinct sd1.slot as s1, sd2.slot as s2
from [dbo].[EXP_REL_SLOT_DOMN] sd1
inner join [dbo].[EXP_REL_SLOT_DOMN] sd2
on sd1.domn = sd2.domn 
and sd1.slot > sd2.slot

--29 seconds
INSERT into tempslot
select distinct sd1.slot as s1, sd2.slot as s2
from [dbo].[EXP_REL_SLOT_DOMN] sd1
inner join [dbo].[EXP_REL_SLOT_DOMN] sd2
on sd1.domn = sd2.domn 
and sd1.slot < sd2.slot

How can i improve the Perfomance of this Insert statements

Thank you

Ashfield3-Jun-09 21:04

3-Jun-09 21:04

Have you had a look at the execution plan? That should show where the most time is taken. Just in passing, why 2 inserts when 1 will do - unless I'm missing something this has the same effect

INSERT into tempslot
select distinct sd1.slot as s1, sd2.slot as s2
from [dbo].[EXP_REL_SLOT_DOMN] sd1
inner join [dbo].[EXP_REL_SLOT_DOMN] sd2
on sd1.domn = sd2.domn 
and sd1.slot != sd2.slot

It may be quicker to use a temp table, take the distinct off this query and do a distinct from the temp table as it may have many less records to process.

Bob
Ashfield Consultants Ltd
Proud to be a 2009 Code Project MVP

Vimalsoft(Pty) Ltd3-Jun-09 21:34

Vimalsoft(Pty) Ltd

3-Jun-09 21:34

Good Morning Ashfield

Thank you for your reply.

Ashfield: wrote
why 2 inserts when 1 will do - unless I'm missing something this has the same effect

If i use <> or != it will take 3:47 Seconds. by separating the Query it took less than that and there thing to note is that there is not Unique key in these table.

Thank you

Vuyiswa Maseko,

Few companies that installed computers to reduce the employment of clerks have realized their expectations.... They now need more and more expensive clerks even though they call them "Developers" or "Programmers."

C#/VB.NET/ASP.NET/SQL7/2000/2005/2008
http://www.vuyiswamaseko.somee.com
http://www.vuyiswamaseko.tiyaneProperties.co.za
vuyiswa@its.co.za
http://www.itsabacus.co.za/itsabacus/

Ashfield4-Jun-09 1:31

4-Jun-09 1:31

Something I just noticed, it would probably benefit to change your index to put on domn then slot. If you think about it, you are matching on domn and looking for slot within in that domn.

Bob
Ashfield Consultants Ltd
Proud to be a 2009 Code Project MVP

smcnulty20004-Jun-09 1:58

smcnulty2000

4-Jun-09 1:58

And consider telling us three more things;

1. Which version sql server, 2005 or 2008?
2. What is the description of tempslot?
3. What is the result when you run the select statements without the insert portion of the statement?

In other words, do they run faster by quite a bit or the same?

_____________________________
Those who study history are doomed to watch others repeat it. -Scott M.

smcnulty20003-Jun-09 22:36

smcnulty2000

3-Jun-09 22:36

Short answer: try removing the index.

Long answer:

How many records are you putting in? That's pretty critical to discussing this. 29 seconds with a billion records is different than 29 seconds with a 1000 records.

You might be better off without the index. I recommend you drop the index and retest your speeds without it.
Realize that indexes are great for lookup but there is a price to pay when you write to a table.

Every time you write to the table, the system has to rebuild the index and since you are using a clustered index you are reordering the data blocks in the same order as the index. So you write, then insert the next line, then reorder, then rewrite the whole mess. That's not exactly efficient.

And because it is a primary key the system will check the uniqueness of each line entered every one of those checks costs you.

Equally important is the fact that you are indexing all columns in one index. You really are just trading disk reads in one place for disk reads in another.

On the other side; when someone runs a select against this table what percentage of the data will be returned on a regular basis? Will it be 5% or less? Or will it be most of the table.

If you aren't going to see 5% or less then having an index is just a waste. Full table scans would be quicker.

With all of that in mind; if you must have the index then see if you can code the situation so that it drops the index, then inserts all the data in one fell swoop, then rebuilds the index.

_____________________________
Those who study history are doomed to watch others repeat it. -Scott M.

Vimalsoft(Pty) Ltd4-Jun-09 0:00

Vimalsoft(Pty) Ltd

4-Jun-09 0:00

Good Morning smcnulty2000

Thanks for your reply

<b>smcnulty2000 wrote :</b>
Short answer: try removing the index.

Without indexes it run 2 Minutes 10 seconds

Thank you for the Advice.

Ashfield4-Jun-09 1:24