Linking a matrix to a chart

Hello,
I do not know if it is possible but let me explain what I want.
I have a matrix which contains to groups on rows: region and process. The
region can be expanded so that the process group becomes visible.
I have created a chart based on the same dataset. What is want is dynamic
grouping based on the visibility of the processgroup of the matrix.
Thus when the report is rendered the matrix only shows data for the region.
When the region is expanded the matrix shows three processes for that region.
I would like that the chart shows the three processes when the region is
expaned. It can be done by changing the grouping value for the chart_series
group from region to process. But I want this to happen dynamically.
Any ideas?
Regards,
QI'm sorry but linking across data regions (e.g. matrix and chart) is not
possible. The closest you can get today is to do this via a report
parameter - i.e. the parameter value determines the show/hide state of the
matrix group and the dynamic grouping expression of the chart.
--
This posting is provided "AS IS" with no warranties, and confers no rights.
"Qbee" <Qbee@.discussions.microsoft.com> wrote in message
news:8C88F103-845C-48BD-A3A5-1DA7C1854015@.microsoft.com...
> Hello,
> I do not know if it is possible but let me explain what I want.
> I have a matrix which contains to groups on rows: region and process. The
> region can be expanded so that the process group becomes visible.
> I have created a chart based on the same dataset. What is want is dynamic
> grouping based on the visibility of the processgroup of the matrix.
> Thus when the report is rendered the matrix only shows data for the
> region.
> When the region is expanded the matrix shows three processes for that
> region.
> I would like that the chart shows the three processes when the region is
> expaned. It can be done by changing the grouping value for the
> chart_series
> group from region to process. But I want this to happen dynamically.
> Any ideas?
> Regards,
> Q|||Thank Robert for your reply.
Is this feature on the whishlist for the a version?
Q
"Robert Bruckner [MSFT]" wrote:
> I'm sorry but linking across data regions (e.g. matrix and chart) is not
> possible. The closest you can get today is to do this via a report
> parameter - i.e. the parameter value determines the show/hide state of the
> matrix group and the dynamic grouping expression of the chart.
> --
> This posting is provided "AS IS" with no warranties, and confers no rights.
>
> "Qbee" <Qbee@.discussions.microsoft.com> wrote in message
> news:8C88F103-845C-48BD-A3A5-1DA7C1854015@.microsoft.com...
> > Hello,
> >
> > I do not know if it is possible but let me explain what I want.
> >
> > I have a matrix which contains to groups on rows: region and process. The
> > region can be expanded so that the process group becomes visible.
> > I have created a chart based on the same dataset. What is want is dynamic
> > grouping based on the visibility of the processgroup of the matrix.
> > Thus when the report is rendered the matrix only shows data for the
> > region.
> > When the region is expanded the matrix shows three processes for that
> > region.
> > I would like that the chart shows the three processes when the region is
> > expaned. It can be done by changing the grouping value for the
> > chart_series
> > group from region to process. But I want this to happen dynamically.
> >
> > Any ideas?
> >
> > Regards,
> > Q
>
>|||Yes, we are aware of the usefulness of this feature and will look into
providing a solution in a future release. The more customers requesting it,
the higher the relative priority compared to many other feature requests.
--
This posting is provided "AS IS" with no warranties, and confers no rights.
"Qbee" <Qbee@.discussions.microsoft.com> wrote in message
news:1ABFFAE3-6C17-4020-856F-6DAE3C588C94@.microsoft.com...
> Thank Robert for your reply.
> Is this feature on the whishlist for the a version?
> Q
> "Robert Bruckner [MSFT]" wrote:
>> I'm sorry but linking across data regions (e.g. matrix and chart) is not
>> possible. The closest you can get today is to do this via a report
>> parameter - i.e. the parameter value determines the show/hide state of
>> the
>> matrix group and the dynamic grouping expression of the chart.
>> --
>> This posting is provided "AS IS" with no warranties, and confers no
>> rights.
>>
>> "Qbee" <Qbee@.discussions.microsoft.com> wrote in message
>> news:8C88F103-845C-48BD-A3A5-1DA7C1854015@.microsoft.com...
>> > Hello,
>> >
>> > I do not know if it is possible but let me explain what I want.
>> >
>> > I have a matrix which contains to groups on rows: region and process.
>> > The
>> > region can be expanded so that the process group becomes visible.
>> > I have created a chart based on the same dataset. What is want is
>> > dynamic
>> > grouping based on the visibility of the processgroup of the matrix.
>> > Thus when the report is rendered the matrix only shows data for the
>> > region.
>> > When the region is expanded the matrix shows three processes for that
>> > region.
>> > I would like that the chart shows the three processes when the region
>> > is
>> > expaned. It can be done by changing the grouping value for the
>> > chart_series
>> > group from region to process. But I want this to happen dynamically.
>> >
>> > Any ideas?
>> >
>> > Regards,
>> > Q
>>

servers and partitioned view?

Hi,
I have 2 database servers running W2K/SQL 2000 with Gigabit Ethernet between
them. Server1 contains 2003 and 2002 data in one database. Server2 contains
2001 data in one database.
I have an web application that points to Server1 all the time. So, I created
a linked server from Server1 to add Server2 in. I then created Server1.VIEWS
of all tables from Server2.database.dbo.table_name...
The problem is that accessing data via Server1.VIEWS from the web
application is very slow. Reports aren't performing for 2001 data
physicially stored on Server2. The reports aggregate millions of rows on
each table.
what are ways to improve query performance via linked servers? I still want
to keep one web application pointing to one physicially server, if possible.
Thanks!
HHIf you haven't done so already, run a trace against the database and try to
index the tables on server2 based on the trace.
For some of the more common reports, if the data is aggregated maybe you can
bypass some of the work SQL Server has to do by creating tables with the
aggregated data in it.
You could try a solution using Analysis Services (this solution would
require a lot of FTE hours)|||Our relational reporting systems works great when the data is local. Reports
usually run within seconds. Only the set of data physically stored on
another machine via linked server performs severely. I didn't expect it to
be this slow because the two machine are on the same Gigabit Ether LAN. It
could be because the query is running on Machine1, but it has to pull all
data from Machine2 over first' It'd be ideal if queries are being run on
Machine2, and only results get transferred over the wire.
Thanks!
HH
"Bianca Blount" <blountb@.noemail.com> wrote in message
news:eZhKrsJmDHA.964@.TK2MSFTNGP10.phx.gbl...
> If you haven't done so already, run a trace against the database and try
to
> index the tables on server2 based on the trace.
> For some of the more common reports, if the data is aggregated maybe you
can
> bypass some of the work SQL Server has to do by creating tables with the
> aggregated data in it.
> You could try a solution using Analysis Services (this solution would
> require a lot of FTE hours)
>

servers

I have created a link between server A and server B
(sever A contains the link info)
I set the default database of the user on server B to be the database I want
to link to on server B.
Inside Enterprise Manager I can view the tables on server B via the link
icon on server A.
When I start Visual Studio and go to the server explorer to design a query
using the GUI interface, when I try to code:
SEVER_NAME.DATABASE.DBO.TABLE or
SERVER_NAME.DBO.TABLE
(Thinking that if the user on the linked servers default database is set,
then no need to specified in the query?)
inside the query it aliases the table name but not columns display in the
design view, just the *. Trying to run the query said that the Name can only
be three parts or invalid object name SERVER_NAME.dbo.TABLE
Like I said I can view the tables via the linking server though EM. My
developers need to be able to use the VS gui to design their queries. I ran
into situations in the past were if I typed the query straight (not using
gui), and try to save it, I would get a message stated that the MS DTC could
not be started, when I know for a fact both server A & B have the DTC runnin
g.
Can anyone provide assistance?SQL Server Service Manager (SQL 2000), and SQL Server Configuration Manager
(SQL 2005).
ML
http://milambda.blogspot.com/|||SQL Server
DTC
and the SQL Ageent are all running.
Where else do I need to look or configure
--
JP
.NET Software Develper
"ML" wrote:

> SQL Server Service Manager (SQL 2000), and SQL Server Configuration Manage
r
> (SQL 2005).
>
> ML
> --
> http://milambda.blogspot.com/

server using the OLEDB for ODBC Driver

I have an ODBC driver to a non-SQL database that contains legacy data that I
need to reference in my application/reports. I had originally hoped to use a
linked server using the OLEDB for ODBC provider to access the data and so be
able to create applications/reports using data stored in SQL Server 2000 and
the legacy database via a single connection.
The problem is that, according to the ODBC drive vendor, when I use the 4
part name to refer to my linked server, SQL Server only passes the driver
the base select statement without any where clause resulting in the entire
table being returned to SQL Server which then applies the filter. This gives
me an average time of 8 minutes to return a single record of a 46,000 row
table.
If I use the OPENQUERY function the same select statement takes about 2
seconds. Unfortunately though, OPENQUERY does not accept a variables as a
parameter and so the select statement must be a hard coded string which
makes it unsuitable for any but a static view.
Any suggestions on a workaround for this?You can build the entire SQL string, e.g. select * from
openquery(server, 'etc...') , and pass the string into an
EXEC(). You can find examples in this article:
HOW TO: Pass a Variable to a Linked Server Query
http://support.microsoft.com/?id=314520
-Sue
On Wed, 11 Aug 2004 12:37:48 -0500, "Charles J Ryan"
<charlesryan1@.msn.com> wrote:

>I have an ODBC driver to a non-SQL database that contains legacy data that
I
>need to reference in my application/reports. I had originally hoped to use
a
>linked server using the OLEDB for ODBC provider to access the data and so b
e
>able to create applications/reports using data stored in SQL Server 2000 an
d
>the legacy database via a single connection.
>The problem is that, according to the ODBC drive vendor, when I use the 4
>part name to refer to my linked server, SQL Server only passes the driver
>the base select statement without any where clause resulting in the entire
>table being returned to SQL Server which then applies the filter. This give
s
>me an average time of 8 minutes to return a single record of a 46,000 row
>table.
>If I use the OPENQUERY function the same select statement takes about 2
>seconds. Unfortunately though, OPENQUERY does not accept a variables as a
>parameter and so the select statement must be a hard coded string which
>makes it unsuitable for any but a static view.
>Any suggestions on a workaround for this?
>

server using the OLEDB for ODBC Driver

I have an ODBC driver to a non-SQL database that contains legacy data that I
need to reference in my application/reports. I had originally hoped to use a
linked server using the OLEDB for ODBC provider to access the data and so be
able to create applications/reports using data stored in SQL Server 2000 and
the legacy database via a single connection.
The problem is that, according to the ODBC drive vendor, when I use the 4
part name to refer to my linked server, SQL Server only passes the driver
the base select statement without any where clause resulting in the entire
table being returned to SQL Server which then applies the filter. This gives
me an average time of 8 minutes to return a single record of a 46,000 row
table.
If I use the OPENQUERY function the same select statement takes about 2
seconds. Unfortunately though, OPENQUERY does not accept a variables as a
parameter and so the select statement must be a hard coded string which
makes it unsuitable for any but a static view.
Any suggestions on a workaround for this?
You can build the entire SQL string, e.g. select * from
openquery(server, 'etc...') , and pass the string into an
EXEC(). You can find examples in this article:
HOW TO: Pass a Variable to a Linked Server Query
http://support.microsoft.com/?id=314520
-Sue
On Wed, 11 Aug 2004 12:37:48 -0500, "Charles J Ryan"
<charlesryan1@.msn.com> wrote:

>I have an ODBC driver to a non-SQL database that contains legacy data that I
>need to reference in my application/reports. I had originally hoped to use a
>linked server using the OLEDB for ODBC provider to access the data and so be
>able to create applications/reports using data stored in SQL Server 2000 and
>the legacy database via a single connection.
>The problem is that, according to the ODBC drive vendor, when I use the 4
>part name to refer to my linked server, SQL Server only passes the driver
>the base select statement without any where clause resulting in the entire
>table being returned to SQL Server which then applies the filter. This gives
>me an average time of 8 minutes to return a single record of a 46,000 row
>table.
>If I use the OPENQUERY function the same select statement takes about 2
>seconds. Unfortunately though, OPENQUERY does not accept a variables as a
>parameter and so the select statement must be a hard coded string which
>makes it unsuitable for any but a static view.
>Any suggestions on a workaround for this?
>