Avoiding time-outs

The C++ application calls the database to look up property data. One
troublesome query is a function that returns a table, finding data which
is assembled from four or five tables through a view that has a join,
and then updating the resulting @.table from some other tables. There
are several queries inside the function, which are selected according
to which parameters are supplied (house #, street, zip, or perhaps parcel
number, or house #, street, town, city,...etc.). If a lot of parameters
are provided, and the property is not in the database, then several queries
may be attempted -- it keeps going until it runs out of queries or finds
something. Usually it takes ~1-2 sec for a hit, but maybe a minute in
some failure cases, depending on the distribution of data. (~100 mil
properties in the DB) Some queires operate on the assumption the input data
is slightly faulty, and take relatively a long time, e.g., if WHERE
ZIP=@.Zip fails, we try WHERE ZIP LIKE substring(@.Zip,1,3)+'%'. While
all this is going on the application may decide the DB is never going to
return, and time out; it also seems more likely to throw an exception the
longer it has to wait. Is there a way to cause the DB function to fail if
it takes more than a certain amount of time? I could also recast it as
a procedure, and check the time consumed after every query, and abandon
the search if a certain amount of time has elapsed.

Thanks in advance,
Jim Geissmanjim_geissman@.countrywide.com (Jim Geissman) wrote in message news:<b84bf9dc.0403031505.2838a043@.posting.google.com>...
> The C++ application calls the database to look up property data. One
> troublesome query is a function that returns a table, finding data which
> is assembled from four or five tables through a view that has a join,
> and then updating the resulting @.table from some other tables. There
> are several queries inside the function, which are selected according
> to which parameters are supplied (house #, street, zip, or perhaps parcel
> number, or house #, street, town, city,...etc.). If a lot of parameters
> are provided, and the property is not in the database, then several queries
> may be attempted -- it keeps going until it runs out of queries or finds
> something. Usually it takes ~1-2 sec for a hit, but maybe a minute in
> some failure cases, depending on the distribution of data. (~100 mil
> properties in the DB) Some queires operate on the assumption the input data
> is slightly faulty, and take relatively a long time, e.g., if WHERE
> ZIP=@.Zip fails, we try WHERE ZIP LIKE substring(@.Zip,1,3)+'%'. While
> all this is going on the application may decide the DB is never going to
> return, and time out; it also seems more likely to throw an exception the
> longer it has to wait. Is there a way to cause the DB function to fail if
> it takes more than a certain amount of time? I could also recast it as
> a procedure, and check the time consumed after every query, and abandon
> the search if a certain amount of time has elapsed.
> Thanks in advance,
> Jim Geissman

You don't give any information about your version of MSSQL, and the
client library you're using, but you may be able to set a suitable
timeout period on the client side. Alternatively, look at the "query
governor cost limit Option" in Books Online - this terminates queries
that run for more than a given number of seconds.

Simon|||jim_geissman@.countrywide.com (Jim Geissman) wrote in message news:<b84bf9dc.0403031505.2838a043@.posting.google.com>...
> The C++ application calls the database to look up property data. One
> troublesome query is a function that returns a table, finding data which
> is assembled from four or five tables through a view that has a join,
> and then updating the resulting @.table from some other tables. There
> are several queries inside the function, which are selected according
> to which parameters are supplied (house #, street, zip, or perhaps parcel
> number, or house #, street, town, city,...etc.). If a lot of parameters
> are provided, and the property is not in the database, then several queries
> may be attempted -- it keeps going until it runs out of queries or finds
> something. Usually it takes ~1-2 sec for a hit, but maybe a minute in
> some failure cases, depending on the distribution of data. (~100 mil
> properties in the DB) Some queires operate on the assumption the input data
> is slightly faulty, and take relatively a long time, e.g., if WHERE
> ZIP=@.Zip fails, we try WHERE ZIP LIKE substring(@.Zip,1,3)+'%'. While
> all this is going on the application may decide the DB is never going to
> return, and time out; it also seems more likely to throw an exception the
> longer it has to wait. Is there a way to cause the DB function to fail if
> it takes more than a certain amount of time? I could also recast it as
> a procedure, and check the time consumed after every query, and abandon
> the search if a certain amount of time has elapsed.
> Thanks in advance,
> Jim Geissman

See "remote query timeout Option" in the help text. However, relying
on this may cause inconsistent bahaviour.

This design pattern can also lead to heavy load on your database.

As a suggestion, have two separate sets of queries, one that assumes
good data (should be much quicker which you want to use most times?)
and one that may have incorrect data (will be slower, but not used
very often). In you screen have a checkbox to indicate what search
option to use. Alternatively perform better validation on the data
before submitting the form.|||That sounds interesting. I'll look into it.

> You don't give any information about your version of MSSQL, and the
> client library you're using, but you may be able to set a suitable
> timeout period on the client side. Alternatively, look at the "query
> governor cost limit Option" in Books Online - this terminates queries
> that run for more than a given number of seconds.
> Simon

Avoid index scan with LIKE and a variable

Hi,
Here's my problem: I want to write a stored procedure that returns all
records from a table that have a certain column starting with given
text. I however find that using LIKE and a variable always causes an
index scan... which is causing performance issues. My table has about
3.5M records.

Below is a test. In query analyser if I look at the execution plan for
the following it will come up as in index scan. However, if i just
hard-code the text it all works fine (index seek).

How can I do this with reasonable speed?
Thanks Greg

DECLARE @.find varchar(50)
SET @.find = 'start'

SELECT TOP 100
*
FROM Test
WHERE
Col1 LIKE @.find + '%'
--Col1 LIKE 'start%'gregbacchus (greg.bacchus@.gmail.com) writes:
> Here's my problem: I want to write a stored procedure that returns all
> records from a table that have a certain column starting with given
> text. I however find that using LIKE and a variable always causes an
> index scan... which is causing performance issues. My table has about
> 3.5M records.
> Below is a test. In query analyser if I look at the execution plan for
> the following it will come up as in index scan. However, if i just
> hard-code the text it all works fine (index seek).
> How can I do this with reasonable speed?
> Thanks Greg
>
> DECLARE @.find varchar(50)
> SET @.find = 'start'
> SELECT TOP 100
> *
> FROM Test
> WHERE
> Col1 LIKE @.find + '%'
> --Col1 LIKE 'start%'

You don't say whether the index on Test.Col1 is clustered or not, and
whether this is the index that is scanned. I would expect that the index
on Test.Col1 is non-clustered, and the scan you see is a clustered index
scan.

When you have the literal, SQL Server knows about the query than you
have the variable. For the variable, SQL Server can only make a standard
assumption. Had the variable instead been a parameter to a stored procedure,
SQL Server would have looked at that value.

The best way out may be to simply use an index hint. You could also use
sp_executesql and pass the variable as a parameter.

--
Erland Sommarskog, SQL Server MVP, esquel@.sommarskog.se

Books Online for SQL Server SP3 at
http://www.microsoft.com/sql/techin.../2000/books.asp

Avoid getting same country twice

I have a DB containing trips around the world.
I want to create a menu with all the countries we visit but at the moment it returns the whole column. i.e. Brazil 5 times etc.
How do I avoid repititions?
I thought I could make an array and check it to see if the country was already there, but I think SQL must have a native routine to achieve the same thing.
Thanks
MFirst of all, you should check your data model and select statement you use to retrieve countries. Probably your SELECT is not defined properly if it returns multiple rows. Anyway, you can retrieve unique names using DISTINCT:

select DISTINCT CountryName
from ...|||cool, thanks er.. madafaka.
my SQL looks like this now, and works a treat:

"SELECT DISTINCT Country, url_link FROM dest_search ORDER BY country"

knowing me there's probably still a better way!
Incidentally the url_link is the path to the detail page and I gather doing it this way allows the search engines to follow the links. at first I put just:
"SELECT DISTINCT Country FROM dest_search ORDER BY country"
but that gave me an error because it excluded the url_link data.

Cheers
M

avg of most current 50 only

I have a query that returns the averages for a selected group of records.
select AVG(h.stkhstClose), h.stkhstcsisym
from stkhst h
JOIN unvmem u on h.stkhstcsisym = u.unvmemCsiId
and u.unvmemUnvID = 29001
group by h.stkhstcsisym
order by h.stkhstcsisym
this works and returns 99 averages. However, I need it to average only the
last 50 records of each group based on the date. I'm looking for a clean way
to do this without using a temporary table or cursor. and ideas would be
appreciated
thanks
kesJust a guess. See my signature for a more precise answer.
SELECT AVG(whatever) FROM (SELECT TOP 50 whatever FROM table WHERE
<whatever> ORDER BY datecolumn DESC) x
Please post DDL, sample data and desired results.
See http://www.aspfaq.com/5006 for info.
"Kurt Schroeder" <KurtSchroeder@.discussions.microsoft.com> wrote in message
news:735FDCDB-B93C-45ED-94EA-0831429F5CB3@.microsoft.com...
> I have a query that returns the averages for a selected group of records.
> select AVG(h.stkhstClose), h.stkhstcsisym
> from stkhst h
> JOIN unvmem u on h.stkhstcsisym = u.unvmemCsiId
> and u.unvmemUnvID = 29001
> group by h.stkhstcsisym
> order by h.stkhstcsisym
> this works and returns 99 averages. However, I need it to average only the
> last 50 records of each group based on the date. I'm looking for a clean
way
> to do this without using a temporary table or cursor. and ideas would be
> appreciated
> thanks
> kes|||Thank You Aaron (you seem to answer a lot of my postings and your suggestion
s
have always proven helpful)
this will get the average for one group but how about the rest? Would a
where stkhstDate IN (select top 50 stkhstdate from stkhst where stkhstid =
xx order by stkhstdate DESC)
thank you
kes
"Aaron [SQL Server MVP]" wrote:

> Just a guess. See my signature for a more precise answer.
> SELECT AVG(whatever) FROM (SELECT TOP 50 whatever FROM table WHERE
> <whatever> ORDER BY datecolumn DESC) x
> --
> Please post DDL, sample data and desired results.
> See http://www.aspfaq.com/5006 for info.
>
>
> "Kurt Schroeder" <KurtSchroeder@.discussions.microsoft.com> wrote in messag
e
> news:735FDCDB-B93C-45ED-94EA-0831429F5CB3@.microsoft.com...
> way
>
>|||>> this will get the average for one group but how about the rest?
Did you read Aaron's post? To repeat:
See his signature for a more precise answer.
Please post DDL, sample data and desired results.
See http://www.aspfaq.com/5006 for info.
Anith|||ok, fair enough.
stkhst:
CREATE TABLE [stkhst] (
[stkhstID] [int] IDENTITY (1, 1) NOT NULL ,
[stkhstCsiSym] [int] NULL ,
[stkhstSym] [varchar] (10) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
[stkhstDate] [int] NULL ,
[stkhstOpen] [decimal](9, 4) NULL ,
[stkhstHi] [decimal](9, 4) NULL ,
[stkhstLow] [decimal](9, 4) NULL ,
[stkhstClose] [decimal](9, 4) NULL ,
[stkhstVol] [int] NULL ,
[stkhstDiv] [int] NULL ,
[stkhstX] [int] NULL ,
[stkhstO] [int] NULL ,
[stkhstXO] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
[stkhstBuySell] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL
CONSTRAINT [DF_stkhst_stkhstBuySell] DEFAULT ('U'),
[stkhstLine] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
[stkhstCPosL] [int] NULL ,
[stkhstCPosH] [int] NULL ,
[stkHstCCol] [int] NULL
) ON [PRIMARY]
GO
unvmem:
CREATE TABLE [unvmem] (
[unvmemRecID] [int] IDENTITY (1, 1) NOT NULL ,
[unvmemCsiId] [int] NOT NULL ,
[unvmemUnvID] [int] NOT NULL ,
[unvmemActive] [bit] NULL CONSTRAINT [DF_unvmem_unvmemActive] DEFAULT (1)
) ON [PRIMARY]
GO
"Aaron [SQL Server MVP]" wrote:

> Just a guess. See my signature for a more precise answer.
> SELECT AVG(whatever) FROM (SELECT TOP 50 whatever FROM table WHERE
> <whatever> ORDER BY datecolumn DESC) x
> --
> Please post DDL, sample data and desired results.
> See http://www.aspfaq.com/5006 for info.
>
>
> "Kurt Schroeder" <KurtSchroeder@.discussions.microsoft.com> wrote in messag
e
> news:735FDCDB-B93C-45ED-94EA-0831429F5CB3@.microsoft.com...
> way
>
>|||noted, posted
thanks
kes
"Anith Sen" wrote:

> Did you read Aaron's post? To repeat:
> See his signature for a more precise answer.
> Please post DDL, sample data and desired results.
> See http://www.aspfaq.com/5006 for info.
> --
> Anith
>
>|||What about sample data and desired results?
The point is that we're not going to drive to Wichita or Kansas or wherever
you are to see what data is in your table and try to figure out what result
you want from that data. And we're certainly not going to spend our
afternoon inventing fictitious but possibly unrealistic data to populate
your empty table, then spend time developing a solution against that, only
to find out all the "buts" that come with the assumptions we made. Please
supply sample data in the form of INSERT statements, and the resultset you
want based on that data.
Please post DDL, sample data and desired results.
See http://www.aspfaq.com/5006 for info.

> ok, fair enough.
> stkhst:
> CREATE TABLE [stkhst] (
> [stkhstID] [int] IDENTITY (1, 1) NOT NULL ,
> [stkhstCsiSym] [int] NULL ,
> [stkhstSym] [varchar] (10) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
> [stkhstDate] [int] NULL ,
> [stkhstOpen] [decimal](9, 4) NULL ,
> [stkhstHi] [decimal](9, 4) NULL ,
> [stkhstLow] [decimal](9, 4) NULL ,
> [stkhstClose] [decimal](9, 4) NULL ,
> [stkhstVol] [int] NULL ,
> [stkhstDiv] [int] NULL ,
> [stkhstX] [int] NULL ,
> [stkhstO] [int] NULL ,
> [stkhstXO] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
> [stkhstBuySell] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL
> CONSTRAINT [DF_stkhst_stkhstBuySell] DEFAULT ('U'),
> [stkhstLine] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
> [stkhstCPosL] [int] NULL ,
> [stkhstCPosH] [int] NULL ,
> [stkHstCCol] [int] NULL
> ) ON [PRIMARY]
> GO
> unvmem:
> CREATE TABLE [unvmem] (
> [unvmemRecID] [int] IDENTITY (1, 1) NOT NULL ,
> [unvmemCsiId] [int] NOT NULL ,
> [unvmemUnvID] [int] NOT NULL ,
> [unvmemActive] [bit] NULL CONSTRAINT [DF_unvmem_unvmemActive] DEFAULT (1)
> ) ON [PRIMARY]
> GO|||Kurt,
Your tables seem to have no primary keys which is a critical design flaw.
Generally for such problems, others cannot test the solutions without sample
data. You have not provided that either. Also as a side note, if your scheme
allows, you may want to look closely at your naming convention as well.
Here is another attempt with guesswork:
SELECT AVG( stkhstClose ), stkhstcsisym
FROM ( SELECT TOP 50 h.stkhstClose, h.stkhstcsisym
FROM stkhst h
INNER JOIN unvmem u
ON h.stkhstcsisym = u.unvmemCsiId
WHERE u.unvmemUnvID = 29001
ORDER BY h.stkhstcsisym ) D ( stkhstClose, stkhstcsisym )
GROUP BY stkhstcsisym ;
Anith|||Kurt,
You want to extract and average the last 50 records for each group...
Just add a where clause that restricts the query to operate only on those
records which have 50 or less "partners" (in the same group) after them...
Select AVG(h.stkhstClose), h.stkhstcsisym
From stkhst h
Where (Select Count(*) From stkhst
Where stkhstcsisym = h.stkhstcsisym
And DateColumn >= h.DateColumn) <= 50
Select * From
"Kurt Schroeder" wrote:
> ok, fair enough.
> stkhst:
> CREATE TABLE [stkhst] (
> [stkhstID] [int] IDENTITY (1, 1) NOT NULL ,
> [stkhstCsiSym] [int] NULL ,
> [stkhstSym] [varchar] (10) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
> [stkhstDate] [int] NULL ,
> [stkhstOpen] [decimal](9, 4) NULL ,
> [stkhstHi] [decimal](9, 4) NULL ,
> [stkhstLow] [decimal](9, 4) NULL ,
> [stkhstClose] [decimal](9, 4) NULL ,
> [stkhstVol] [int] NULL ,
> [stkhstDiv] [int] NULL ,
> [stkhstX] [int] NULL ,
> [stkhstO] [int] NULL ,
> [stkhstXO] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
> [stkhstBuySell] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL
> CONSTRAINT [DF_stkhst_stkhstBuySell] DEFAULT ('U'),
> [stkhstLine] [char] (1) COLLATE SQL_Latin1_General_CP1_CI_AS NULL ,
> [stkhstCPosL] [int] NULL ,
> [stkhstCPosH] [int] NULL ,
> [stkHstCCol] [int] NULL
> ) ON [PRIMARY]
> GO
> unvmem:
> CREATE TABLE [unvmem] (
> [unvmemRecID] [int] IDENTITY (1, 1) NOT NULL ,
> [unvmemCsiId] [int] NOT NULL ,
> [unvmemUnvID] [int] NOT NULL ,
> [unvmemActive] [bit] NULL CONSTRAINT [DF_unvmem_unvmemActive] DEFAULT (1)
> ) ON [PRIMARY]
> GO
> "Aaron [SQL Server MVP]" wrote:
>|||My apologies, I did not mean to imply that I needed more than advise. Please
understand that I do not feel it appropriate to ask for more than just that.
I feel it would be unfair to you or anyone else to do my work for me.
Aaron, your first posting to my question gave me what I needed to search for
the answer. My real query is much more complex, but this part of it was
simple enough to post for advice.
Again I wish to thank you for your help.
Humbly yours
kes
"Aaron [SQL Server MVP]" wrote:

> What about sample data and desired results?
> The point is that we're not going to drive to Wichita or Kansas or whereve
r
> you are to see what data is in your table and try to figure out what resul
t
> you want from that data. And we're certainly not going to spend our
> afternoon inventing fictitious but possibly unrealistic data to populate
> your empty table, then spend time developing a solution against that, only
> to find out all the "buts" that come with the assumptions we made. Please
> supply sample data in the form of INSERT statements, and the resultset you
> want based on that data.
> --
> Please post DDL, sample data and desired results.
> See http://www.aspfaq.com/5006 for info.
>
>
>
>

AVG of integer column returns integers result - newbie

I am averaging integer columns, the result is integer not real or decimal
How do I get aroung this?
SELECT AVG(AmpReading)
FROM tblChillWaterSystems
GROUP BY LocationID, SystemID
Values are
1
3
6
--
10
10/3 = 3.33 but the result is 3SELECT AVG(CAST(AmpReading AS REAL)) ...
David Portas
SQL Server MVP
--|||Try,
SELECT AVG(AmpReading * 1.00)
FROM tblChillWaterSystems
GROUP BY LocationID, SystemID
or cast [AmpReading] to numeric.
AMB
"Craig" wrote:

> I am averaging integer columns, the result is integer not real or decimal
> How do I get aroung this?
> SELECT AVG(AmpReading)
> FROM tblChillWaterSystems
> GROUP BY LocationID, SystemID
> Values are
> 1
> 3
> 6
> --
> 10
> 10/3 = 3.33 but the result is 3
>
>

AVG of integer column returns integers result - newbie

I am averaging integer columns, the result is integer not real or decimal
How do I get aroung this?
SELECT AVG(AmpReading)
FROM tblChillWaterSystems
GROUP BY LocationID, SystemID
Values are
1
3
6
10
10/3 = 3.33 but the result is 3
SELECT AVG(CAST(AmpReading AS REAL)) ...
David Portas
SQL Server MVP
|||Try,
SELECT AVG(AmpReading * 1.00)
FROM tblChillWaterSystems
GROUP BY LocationID, SystemID
or cast [AmpReading] to numeric.
AMB
"Craig" wrote:

> I am averaging integer columns, the result is integer not real or decimal
> How do I get aroung this?
> SELECT AVG(AmpReading)
> FROM tblChillWaterSystems
> GROUP BY LocationID, SystemID
> Values are
> 1
> 3
> 6
> --
> 10
> 10/3 = 3.33 but the result is 3
>
>

AVG of integer column returns integers result - newbie

I am averaging integer columns, the result is integer not real or decimal
How do I get aroung this?
SELECT AVG(AmpReading)
FROM tblChillWaterSystems
GROUP BY LocationID, SystemID
Values are
1
3
6
--
10
10/3 = 3.33 but the result is 3SELECT AVG(CAST(AmpReading AS REAL)) ...
--
David Portas
SQL Server MVP
--|||Try,
SELECT AVG(AmpReading * 1.00)
FROM tblChillWaterSystems
GROUP BY LocationID, SystemID
or cast [AmpReading] to numeric.
AMB
"Craig" wrote:
> I am averaging integer columns, the result is integer not real or decimal
> How do I get aroung this?
> SELECT AVG(AmpReading)
> FROM tblChillWaterSystems
> GROUP BY LocationID, SystemID
> Values are
> 1
> 3
> 6
> --
> 10
> 10/3 = 3.33 but the result is 3
>
>

Avg function

The following function returns an average in minutes between to dates.
=Avg(Time.ClsReportUtils.getResponseTime(Fields!CREATED.Value,
Fields!TODO_ACTL_END_DT.Value ))/60
My problem is in figuring out how to truncate decimal places for example the
report returns 36.2795358649783 when all I want is 36.
If any one has any ideas that would be greatly appreciated.
--
kmatth007use format(Avg(Time.ClsReportUtils.getResponseTime(Fields!CREATED.Value,
> Fields!TODO_ACTL_END_DT.Value ))/60,"0")
"kmatth007" wrote:
> The following function returns an average in minutes between to dates.
> =Avg(Time.ClsReportUtils.getResponseTime(Fields!CREATED.Value,
> Fields!TODO_ACTL_END_DT.Value ))/60
> My problem is in figuring out how to truncate decimal places for example the
> report returns 36.2795358649783 when all I want is 36.
> If any one has any ideas that would be greatly appreciated.
> --
> kmatth007|||Thanks! This worked.
kmatth007
"ש×?×?×?" wrote:
> use format(Avg(Time.ClsReportUtils.getResponseTime(Fields!CREATED.Value,
> > Fields!TODO_ACTL_END_DT.Value ))/60,"0")
> "kmatth007" wrote:
> > The following function returns an average in minutes between to dates.
> >
> > =Avg(Time.ClsReportUtils.getResponseTime(Fields!CREATED.Value,
> > Fields!TODO_ACTL_END_DT.Value ))/60
> >
> > My problem is in figuring out how to truncate decimal places for example the
> > report returns 36.2795358649783 when all I want is 36.
> >
> > If any one has any ideas that would be greatly appreciated.
> >
> > --
> > kmatth007

average question

SELECT AVG(SCORE) AS AVERAGE_SCORE
FROM GRADES
SCORE is an integar column with values from 0 - 5
Now this only returns integer values such as 3, 4...
How can I make AVERAGE_SCORE to be a decimal?
ie 3.452
Howardselect avg(1.0*SCORE) AS AVERAGE_SCORE FROM GRADES
Without the 1.0*, the average will be SUM(SCORE)/COUNT(SCORE),
which will be a quotient of integers, and use integer division which
discards any remainder.
Steve Kass
Drew University
"Howard" <howdy0909@.yahoo.com> wrote in message
news:uxcoNcggGHA.3996@.TK2MSFTNGP03.phx.gbl...
> SELECT AVG(SCORE) AS AVERAGE_SCORE
> FROM GRADES
> SCORE is an integar column with values from 0 - 5
> Now this only returns integer values such as 3, 4...
> How can I make AVERAGE_SCORE to be a decimal?
> ie 3.452
>
> Howard
>|||Thanks Steve
One more question
Is it possible to update the value of the field CLASS_AVG in the same query?
I tried this but it didn't work
UPDATE RESULTS
SET CLASS_AVG = AVERAGE_SCORE IN
(SELECT AVG(SCORE) AS AVERAGE_SCORE
FROM GRADES)
"Steve Kass" <skass@.drew.edu> wrote in message
news:u99HpjggGHA.1264@.TK2MSFTNGP05.phx.gbl...
> select avg(1.0*SCORE) AS AVERAGE_SCORE FROM GRADES
> Without the 1.0*, the average will be SUM(SCORE)/COUNT(SCORE),
> which will be a quotient of integers, and use integer division which
> discards any remainder.
> Steve Kass
> Drew University
> "Howard" <howdy0909@.yahoo.com> wrote in message
> news:uxcoNcggGHA.3996@.TK2MSFTNGP03.phx.gbl...
>|||If you are sure of the functioanltiy then may be you should try this.
UPDATE RESULTS
SET CLASS_AVG = (SELECT AVG(SCORE)
FROM GRADES)
But remember this will update the class_avg with the average that you
caclulate for all the rows in the table.|||Howard (howdy0909@.yahoo.com) writes:
> One more question
> Is it possible to update the value of the field CLASS_AVG in the same
> query?
> I tried this but it didn't work
> UPDATE RESULTS
> SET CLASS_AVG = AVERAGE_SCORE IN
> (SELECT AVG(SCORE) AS AVERAGE_SCORE
> FROM GRADES)
You can say simply:
UPDATE RESULTS
SET CLASS_AVG = (SELECT AVG(SCORE) AS AVERAGE_SCORE FROM GRADES)
But this would update every row in RESULTS with the same value,
which may not be what you want.
Erland Sommarskog, SQL Server MVP, esquel@.sommarskog.se
Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/pr...oads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodin...ions/books.mspx|||Just put the query in a VIEW and it will be re-calculated each time.
You are still thinking like a COBOL programmer who wants to write all
his data to a file, not a like an SQL programmer who knows that a VIEW
is also a TABLE and does not have to havea physical existence.|||Or use
select cast(score as decimal (3,2)) as average_score
Same effect, but the explicit cast may be clearer to the next person who
review the code. The next programmer may not know that 1.0*score actually
converts it to a decimal from an int. I never understood it until I started
following this newsgroup.
"Steve Kass" <skass@.drew.edu> wrote in message
news:u99HpjggGHA.1264@.TK2MSFTNGP05.phx.gbl...
> select avg(1.0*SCORE) AS AVERAGE_SCORE FROM GRADES
> Without the 1.0*, the average will be SUM(SCORE)/COUNT(SCORE),
> which will be a quotient of integers, and use integer division which
> discards any remainder.
> Steve Kass
> Drew University
> "Howard" <howdy0909@.yahoo.com> wrote in message
> news:uxcoNcggGHA.3996@.TK2MSFTNGP03.phx.gbl...
>

Avareges for period

Hi,

I am trying to get average for sales for last 30 days.

Aggregation function for [Measures].[Avg Sales] is AverageOfChildren

Next code returns correct result

SELECT { [Measures].[Avg Sales] } ON columns

FROM [Sales]

where ([Dim Date].[Year - Day].[Day].&[2/14/2007]&[2]:[Dim Date].[Year - Day].[Day].&[3/15/2007]&[3])

But this query is not flexible. Should be selected day and returned value for 30 days back.

For example, I am trying to use next code

select

([Dim Date].[Year - Day].[Day].CurrentMember.lead(29) : [Dim Date].[Year - Day].[Day].CurrentMember

, [Measures].[Avg Sales]) // average for 30 days back

ON columns

FROM [SAles]

where ([Dim Date].[Year - Day].[Day].&[3/15/2007]&[3])

But recieved error "The Tuple function expects a tuple expression for the argument. A tuple set expression was used."

What the query should be to resolve this problem?

Thanks for help.

Try creating a query calculated measure for this,like:

With

Member [Measures].[Trailing30Sales] as

Aggregate({[Dim Date].[Year - Day].Lag(29) : [Dim Date].[Year - Day].CurrentMember},

[Measures].[Avg Sales])

select

{[Measures].[Trailing30Sales]} // average for 30 days back

ON columns

FROM [Sales]

where ([Dim Date].[Year - Day].[Day].&[3/15/2007]&[3])

|||

Thank you, very much!

It is very simple