Reports with 7000 records taking 10mts

Hi
I have designed a IR report which can retrieve between 10 records to 500,000 records. But for retrieving 7000 records it tooks 10 minutes.
Could anyone tell me how to improve the performance.

MUnky
thanks for your response. I have recreated a interactive report.
I enabled 10046 trace and i run the report with 15 row at 1st time after i changed the interactive rows per page 500 records and navigated it twice. My trace file shows 15 seconds overall but the apex debug session took 45 seconds. I have create some partition on that table. Table has 4.8 million rows and size is 1.5gig.
The following are the trace output
select
ROWID as apxws_row_pk,
"APXSEQ",
"JOB_SEQ_ID",
"ISSUE_ID",
"TARGET_TABLE_NAME",
"PROGRAM_NAME",
"ISSUE_COL_NAME",
"ISSUE_COL_VALUE",
"ISSUE_TYPE",
"KEY_COL_NAME_1",
"KEY_COL_VALUE_1",
"KEY_COL_NAME_2",
"KEY_COL_VALUE_2",
"KEY_COL_NAME_3",
"KEY_COL_VALUE_3",
"KEY_COL_NAME_4",
"KEY_COL_VALUE_4",
"KEY_COL_NAME_5",
"KEY_COL_VALUE_5",
"KEY_COL_NAME_6",
"KEY_COL_VALUE_6",
"KEY_COL_NAME_7",
"KEY_COL_VALUE_7",
"KEY_COL_NAME_8",
"KEY_COL_VALUE_8",
"KEY_COL_NAME_9",
"KEY_COL_VALUE_9",
"KEY_COL_NAME_10",
"KEY_COL_VALUE_10",
"KEY_COL_NAME_11",
"KEY_COL_VALUE_11",
"KEY_COL_NAME_12",
"KEY_COL_VALUE_12",
"KEY_COL_NAME_13",
"KEY_COL_VALUE_13",
"KEY_COL_NAME_14",
"KEY_COL_VALUE_14",
"KEY_COL_NAME_15",
"KEY_COL_VALUE_15",
"ORA_ERROR_MESSAGE",
"RUN_DATE",
"BLOCK_NAME",
count(*) over () as apxws_row_cnt
from (
select * from (
select * from obe.APX_EXCpt_bkp1 where apxseq=3 and issue_id='102.21'
) r
) r where rownum <= to_number(:APXWS_MAX_ROW_CNT)
order by ROWID
call count cpu elapsed disk query current rows
Parse 3 0.00 0.00 0 0 0 0
Execute 3 0.00 0.00 0 0 0 0
Fetch 1806 0.15 0.14 0 1473 0 1806
total 1812 0.15 0.14 0 1473 0 1806
Misses in library cache during parse: 0
Optimizer mode: ALL_ROWS
Parsing user id: 478 (OBE) (recursive depth: 1)
Rows Row Source Operation
202 WINDOW SORT (cr=491 pr=0 pw=0 time=42548 us)
7039 COUNT STOPKEY (cr=491 pr=0 pw=0 time=91640 us)
7039 PARTITION RANGE SINGLE PARTITION: 3 3 (cr=491 pr=0 pw=0 time=77555 us)
7039 PARTITION HASH SINGLE PARTITION: 2 2 (cr=491 pr=0 pw=0 time=63451 us)
7039 TABLE ACCESS BY LOCAL INDEX ROWID APX_EXCPT_BKP1 PARTITION: 18 18 (cr=491 pr=0 pw=0 time=49367 us)
7039 INDEX RANGE SCAN APX_EXCPT_BKP1_IDX1 PARTITION: 18 18 (cr=21 pr=0 pw=0 time=7174 us)(object id 1059619)
Rows Execution Plan
0 SELECT STATEMENT MODE: ALL_ROWS
202 WINDOW (SORT)
7039 COUNT (STOPKEY)
7039 PARTITION RANGE (SINGLE) PARTITION: START=3 STOP=3
7039 PARTITION HASH (SINGLE) PARTITION: START=2 STOP=2
7039 TABLE ACCESS MODE: ANALYZED (BY LOCAL INDEX ROWID) OF
'APX_EXCPT_BKP1' (TABLE) PARTITION: START=18 STOP=18
7039 INDEX MODE: ANALYZED (RANGE SCAN) OF
'APX_EXCPT_BKP1_IDX1' (INDEX) PARTITION: START=18 STOP=18
Trace file shows fetch count is equal to rows count. is there any option in apex to do bulk fetch?
And munky what is the tuned response time to retrieve 1000 records per page and navigation to next page time

Similar Messages

Date based report with multiple record return

Hello all,
Here is the situation. Running MAS 4.1 and using Crystal 10 for doing reports. I currently have a custom report that shows all invoices that have been paid for any given date range. This works great except when there is a "credit Memo" against an invoice. The way I'm getting all paid invoices is via a formula. The record set I have to pull against lists the invoice amount, date paid and amount paid. Simple calculation on the returned record will tell you if it has been paid off or not. The problem I'm having is those invoice records that also have a credit memo against them. The credit memo entries do not have a "paid date" and since that is my main selection criteria those records will not pull into my report, and therefor those invoices will not be showed as being paid. What I need to happen (and I don't know how to do it). Is for the report to run based upon the date range specified, and pull in related records even if there is no "paid date" on them. The way MAS stores it's invoice/credit memo records is the same for both. The main difference is that invoice has the "type" field set to "IN" and a credit memo has the "type" field set to "CM". So the actual "invoice number" is the same for both kinds of records. I hope my request makes sense, if not let me know and I will try to explain what does not make sense. Thanks in advance. Bill

Jason,
You've been very helpful, but I guess I'm just not explaining myself that well. The Credit Memo's throw things off because they do not contain a "Paid Date" value. That field in the record is blank. Since it is blank the report query ignores them. Therefore when the report query returns all records that have a "Paid date" range of "x to y" they are never returned. This is a problem whenever a credit memo exists for an invoice because the "invoice total" never changes. So when a customer pays on their account for a particular invoice, they only pay what is due which is the "invoice total" minus any "credit" from the "Credit Memo". So when my A/R person receives payment they are showing the invoice paid in full because MAS automatically includes all invoices and credit memo's. My report does not because there is no data in the "Paid Date" field of the record for Credit Memos.
I've been doing alot research in trying to resolve this "exception" to my report. I think I may be on to something, but my knowledge of arrays is extremely limited. My resolution involves populating an array with the "invoice numbers" that are returned from the initial query from the supplied date range. Then using the "invoice numbers" from the array have it then populate my details section of the report with all of "invoice records" both the actual invoice record and any "credit memo" records. I can then group the records returned based upon the "invoice number". then within that grouping I can performing my calculations to show whether or not the invoice has been paid in full. What do you think? I've been able to populate an array with the invoice numbers based upon my date search range, but I have yet to figure out how to take that information and perform another lookup to pull in the rest of the data I need. Do you have any ideas?
Thanks. Bill

Report with 65K records

Hello,
Report outputs 65K records if no selection is made on the variable screen. System generates time out.
I will reduce the number of records by only one plant selection in thevariable screen.
Are there any limitations to the numbers of records in the reports.
Thanks

Hi,
If you are using BW3.5 system , the report will open in analyzer that is excel sheet which can hold only 65,535 records.
In BI7.0 we can hold more than 1 million.
If system is throwing error like timed out, it may comes under performance tuning issues.
We can do it by 1) creating aggregates
2) partitioning cubes....etc..
Hope it helps u,
Thanks,
Sai Chand.S

Speed Issue in Report With 1130680 Records Returned

Hi
<b>Apologies to the mods if this is in the wrong section, it contains content related to Legacy SDKs (VB6/COM), newer .NET content as well as report design issue.</b>
I'm using Windows XP SP3 with the database hosted on my machine, an [email protected] with 2GB DDR2 RAM. Neither the client's pc nor their server are as powerful as mine. (It sucks, I know.)
A report has a speed issue for the end users. The query to the database (SQLS2008) takes 30 seconds and returns 1 130 680 records. And yes, this is the amount of records that need to be returned (the report is a Sales Details report for transactions from 2008/09/01 to 2009/03/31) Running the query through SQLSMS takes the same amount of time whether I execute the statement that the report sends or put the query into a stored procedure. The relevant indices are in place.
Our application is written in VB6 and uses the CR9 unmanaged runtime. I've written a small little CR application in VS2008 to see if running the report through the CR.NET runtime would make it quicker. I built the .NET app in release mode (with optimization and no debug data) and ran the exe to do the speed tests. The times taken for the report to run:<ul>
<li>VB6 app - 3:30s
<li>.NET app - 4:30s
<li>VS2008 IDE - 2:02s
</ul>
I have run the report three times for each app using my cellphone as a stopwatch and the times are reproducible to within two seconds.
I accidentally ran the report in the VS2008 IDE and it took 2:02s to run. Thinking that the loading of the DLLs in .NET was part of the problem, I added code under a button on the form with the following code:
CrystalReportsApplication1.Properties.Settings a = new CrystalReportsApplication1.Properties.Settings();
Assembly aAss = Assembly.LoadFile(a.CR01Engine);
Assembly bAss = Assembly.LoadFile(a.CR02EntFramework);
Assembly cAss = Assembly.LoadFile(a.CR03EntInfoStore);
Assembly dAss = Assembly.LoadFile(a.CR04ReportSource);
Assembly eAss = Assembly.LoadFile(a.CR05Shared);
Assembly fAss = Assembly.LoadFile(a.CR06WinForm);
to preload the assemblies. Unfortunately it made virtually no difference to the running time of the report.
Suggestion from The specified item was not found. on [this page|Speed issue - internal processing vs views;:
8. For summaries, use conditional formulas instead of running totals when possible.
I didn't know that, but we're doing that already.
I even tried using ngen in a virtual machine for the .NET app, but the time was the same. And yes, I know that ngen is only supposed to aid startup times and that it is actually recommended to let the JIT optimize on the fly instead of ngen optimize beforehand. What really struck me was that even in a VM (using Virtual PC 2007) using 512MB of RAM and only a single core the time was the same.
My questions:<ol>
<li>I thought that VS2008 shipped with CR2008 Basic, but according the the references in my project it's CR10. Would the CR2008 runtimes improve the report speed?
<li>Why do the CR10 .NET runtimes take longer to execute than the CR9 COM runtimes?
<li>I really, really, really need the .NET runtime to outperform the COM runtimes as our purchase of CR2008 hinges on this. As you can probably guess, we're currently using ye olde CR9.2 from 2002.
<li>Why oh why do the CR10 .NET runtimes take longer to execute than the CR9 COM runtimes?
<li>The first level of grouping is a variable which changed the grouping depending on a parameter. I changed it to a field from one of the tables and the times for the apps are:
<ul>
<li>VB6 - 2:15s
<li>.NET - 2:48s
<li>CR9.2 IDE - 2:19s
<li>VS2008 IDE - 2:34s
</ul>I may or may not be able to push it out to the client like this, I don't know. Why did the VS2008 IDE take longer this time? All I did was change the grouping-on-parameter to ordinary grouping.
<li>Why oh why oh WHY do the CR10 .NET runtimes take longer to execute than the CR9 COM runtimes? Are they just thunkers to the COM DLLs?
<li>Why does the run in the VS2008 IDE (in debug mode BTW, not release mode) run quicker than both programs?
<li><i>AndrewJ.requestACookie(Flavour.Strawberry, Flavour.Alternate.ChocChip, RequestMethod.Please); // Hey, it's 5:44PM and I want to go home for the long weekend! </i>
</ol>
Thank you and have a blessed Easter!
Edited by: AndrewJ on Apr 14, 2009 10:04 AM. Changed the CPU, I was thinking of my CPU at home.

Please re-post if this is still an issue to the .NET Development - Crystal Reports Forum or purchase a case and have a dedicated support engineer work with you directly.
Bottom line is CR Basic is version 10.5. CR 2008 is 12.2.x.xxxx. Install CR 2008 SP 1 to get .NET support.
And you can't compare COM to .NET, it's like comparing apples and oranges, both fruits but different flavors.

How to create and process Iinteractive report with selected records in 4.1

Hi,
I want to have an Interactive report that will have in front of each row a checkbox that can be selected for batch processing. Is there a way to do so?
If there is please include some method to process those records i.e. for each selected record create a new record on a different table and update the record itself as processed.
thanks
Edited by: yannisr on Jul 13, 2012 5:58 AM

For this you need to consult APEX documentation for two things
Create a checkboxes using APEX_ITEM API
How to reference those checkboxes within an On Submit Process

Report with spreadsheet as desformat opens very slow

Hi All,
Greetings.
We have an xls file which was generated by report runtime.
This xls was created with option desformat=spreadsheet in report
runtime. The problem is it takes considerably long time to open.
More than 5 minutes for a file with 6750 records.
-------scenario
create a rdf report file, say test.rdf.
run test.rdf using SRW.run_report(myPlist)
with option for desformat=spreadsheet.
and DESNAME=abc.xls
the report runs fine and creates an xls file with 6700 records.
open excel application.
open abc.xls
opening abc.xls is painfully slow.
any ideas/suggestion?
thanks in advnace,
-Jafer

hi all,'
Sadly oracle suggested that,
spreadsheet format is not to be used for running reports with large records,
its meant to be used only for small reports.
The reason is report server adds tags to xls report generated using spreadsheet format.
There is a workaround for this issue, if opened in excel application and saved as
excel , the new file looses all the tags but keeps the format.
so it reduces to 1/5th of acutal size and opens quick.
cheers,
-J

Report with all Data Targets in Production and Record Count

Hi,
I am planning to create new report to handle some production maintenance work.
How do I create a report with list of each ODS and Cube and count of Active record count? Can I use any report in BW statistics and modify based on my requirements?
And also "Last Activation Date" is required.
Any suggestions? Please do let me know.
Thanks
Anand.

Hi,
The load might be taking too mucjh time as the system is busy.
wait for some time (30 mins) and see if the load goes thru
The load might have strucked up.
Change the request to red and reload data.
after doing initwith out data load, Run Delta.
Cheers,
Srinath.

Showing Duplicate records in a report with the count of their occurrence

Hi Members,
I am novice to BI. i need your suggestion to achiever one of the functionality. A report is required to be built with columns Incident ID, Task ID, Task Name. The Original report is below:-
In the above report I have to display only duplicated records once. Distincat records are not required to be shown. That is, if the same no. of Task Name is associated with the same Incident ID then it should be displayed once with the count of their occurrences in a separate column say , Count. For example, the Task 'TASK_MANUAL_KCI' got 4 times associated with the same Incident ID INC000000001434. Then the report should display this task one time in a report with count 4. Similarly, for INC000000000943 where task 'IPCG Diagnostic Template' got associated 2 times. The Count is 2 for this task.
Other records should not be displayed in a report.
Would highly appreciate quick response. Please suggest.
Thanks,
Neha Pateria

I tried this Gill, but result is bit different :-
Seems some little modifications are needed. Report should be displayed like the below one where it says how many times the same Task is associated with the same incident. Eg. Task 'TASK_CIRCUIT_RESOLVER' is associated with the incident 'INC000000001434' 4 times. 'IPCG Diagnostic Template' is associated with 'INC000000000943' 2 times. Similarly for other records.
But I really thankful to you for giving me the logic proceed further. I tried '=RunningCount([Task Name]; Row; ([Task Name]; [Incident ID]))'. It gave me desired results. :-
But the only thing that needs to be done is to bring the single Task associated with the Incident..
Thanks,
Neha Pateria

Report rdf with size 8mb taking more time to open

Hello All,
I have a rdf ( reports 6i) report with size 8.5mb taking more time to open and more time to access each field.
Please let me know how do i solve this issue.
Please do the needful.
Thanks.

Thanks for immediate response.
pls let me know how do i know this.
Right now i have the below details from report->help
Report Builder 6.0.8.11.3
ORACLE Server Release 8.0.6.0.0
Oracle Procedure Builder 6.0.8.11.0
Oracle ORACLE PL/SQL V8.0.6.0.0 - Production
Oracle CORE Version 4.0.6.0.0 - Production
Oracle Tools Integration Services 6.0.8.10.2
Oracle Tools Common Area 6.0.5.32.1
Oracle Toolkit 2 for Windows 32-bit platforms 6.0.5.35.0
Resource Object Store 6.0.5.0.1
Oracle Help 6.0.5.35.0
Oracle Sqlmgr 6.0.8.11.3
Oracle Query Builder 6.0.7.0.0 - Production
PL/SQL Editor (c) WinMain Software (www.winmain.com), v1.0 (Production)
Oracle ZRC 6.0.8.11.3
Oracle Express 6.0.8.3.5
Oracle XML Parser 1.0.2.1.0 Production
Oracle Virtual Graphics System 6.0.5.35.0
Oracle Image 6.0.5.34.0
Oracle Multimedia Widget 6.0.5.34.0
Oracle Tools GUI Utilities 6.0.5.35.0
Thanks
Edited by: Abdul Khan on Jan 26, 2010 11:54 PM

Report Painter: Possible to show Report with Zero (not No records were sel)

Report Painter: Possible to show Report with Zero (not No records were selected)
Dear Guru,
is there any way to make Report Painter to show Report with Zero / 0 when there were no record were found. (No records were selected).
Regards,
Howard

hummm, I dont think so
THats not possible, as far as I know
Regards
Ajay M

How to inlcude all records in a report, with a formula result of 0 or 0?

I'm reporting on classes scheduled for a certain location, and calculating the number of available seats per class. A formula will give me the net seats available (Max seats-count of students). BUT if there are no students enrolled, class does not appear on the report. Formula is {SCHEDULES.qy_sch_max} - Count ({WAITING_LISTS.no_emp}, {SCHEDULES.cd_crs)
What am I missing? Currently using Crystal 8.5 and reporting against a SQL database.
Appreciate your help.
christi

A: How to inlcude all records in a report, with a formula result of 0 or >0?

thanks, Sanjay - been there done that.
basically, I need to show the net result for the open seats available,which will be the same as the max seats offered.
I tried the following:
If Isnull{WAITING_LISTS.no_emp}
then {@Open}={SCHEDULES.qy_sch_max}
else
{SCHEDULES.qy_sch_max} - Count ({WAITING_LISTS.no_emp}, {SCHEDULES.cd_crs})
and got a message "Missing the then". (@Open is the formula title)

Apex report with details below for one record

Hi,
I'm looking for an answer to my problem:
We have a report with many columns ~60.
And is not very user friendly because it is so wide table.
If it is possible to display some less important columns like additional information below report,
for single record in report. E.g. user selects radio button from first column and details are refreshed for that record.
(if it is possible to group details in tabs)
Thank you for your answers.

Hi,
Here is one example
https://apex.oracle.com/pls/apex/apex/f?p=39006:41
Create hidden item that is not protected. In my case item name is P41_DEPTNO
First report query
select
APEX_ITEM.RADIOGROUP(1,DEPTNO,:P41_DEPTNO) AS rg,
EMPNO,
ENAME,
JOB,
MGR,
HIREDATE,
SAL,
COMM,
DEPTNO
FROM EMP
Second report query
select LOC,
DNAME,
DEPTNO
from DEPT
WHERE DEPTNO = :P41_DEPTNO
In second report region set Page Items to Submit your hidden item name.
Create dynamic action
- event: Change
- Selection Type: jQuery Selector
- jQuery selector: input[name=f01]
- Action: Set Value
- Set Type: JavaScript Expression
- JavaScript Expression: this.triggeringElement.value;
- Selection Type: Item(s)
- Item(s): <Your hidden item name>
Then add another true event to dynamic action
- Action: Refresh
- Selection Type: Region
- Region: <select second report region>
Regards,
Jari

Access Web Database - Select record and make report with all associated records

Hey everyone,
Right now I'm in the middle of trying to convert an Access client database to be web compatible and I'm running into some problems. For this question, I think I may need to explain a little bit about the database:
The database I'm making is designed to store information about music rights for different songs. The users can input information about writers, producers, organizations, properties, businesses - which is stored all in different related tables. Then, when a
user inputs a song, they choose which writers, produces, organizations, etc. are affiliated with that song.
What I'm trying to do is make a report where you can choose a writer from the list of all the writers and then produce a report with all of the songs by that writer.
I was able to do this in the Access client by making a report that, when opened would trigger (using the On Open event) a form to open where you would choose a writer from a combo box and then click a button. When the button was clicked, it would use the value
in the combo box in a query, which would find all of the songs by that writer and then open up the report which would have the writer and all of their songs on it.
Because web reports don't have many event options and web queries are very limited, I have not found a way to make this report.
Any help at all would be greatly appreciated!

Hi,
I found that you've cross post the quesion on our Answer forum, are you satisfiled the reply from there?
http://answers.microsoft.com/en-us/office/forum/office_2010-access/web-database-select-record-and-make-report-with/04ce4e25-a964-4146-9a34-f9cb26bb0496
Regards,
George Zhao
TechNet Community Support

Crystal Report with 2 oracle datasources (left outer join) very slow

I've made a crystal report with 2 oracle datasources (2 commands). I'm using crystal 10.
These 2 data sources are linked with a left outer join.
The report takes a while to run (more then one hour).
i can run Both query's in a couple of seconds/minutes, but it looks like crystal is runniing the second query for each record in the first query.
When i make the same report in BO. Just 2 queries with merged dimensions in the report, it is taking a couple of minutes to complete the report.
Question is if somebody knows how crystal is handling these 2 different data sources.
Is there any way to say to crystal to fetch the data of both queries and do the join after that?
At the moment it looks like that crystal is going to the other datasource for each record in the first query, which will cost a lot of time.

Joris,
I've always had a bad time combining a Command with any other object. Performance seems to drop dramatically, just as you've described.
I can't tell you specifically why, it falls off so bad...
The solution I've used is to do a linked server query (at least that's what it's called in MS SQL Server) I've never used Oracle, but I'd be VERY surprised if it didn't have that same feature. This will keep 100% of the processing on the server(s) and will get your run times back to what you would expect.
HTH,
Jason

Purchase order report with basic price and Excise duty values

Hi All,
Is there any standard report in SAP to get the Purchase order basic price / qty and its Excise duty values ? Since we are including Excise duties for non - codified items (w/o material master and with cost center 'K') Purchase order and the ed values are added to the cost and the requirement is to see the split up (Basic priceEdEcs+Hcs)for these Purchase orders for the vendors. If there is no std report then please let me know how to link Purchase order table with this Condition value table in query to get the desired report.
Thanks in advance
Benny

Hi,
there is no standert report to fullfill this requierment
you have to create this report with the help of abaer
for that which tax code you are using its importnat thing
for po table EKPO
1. A524 (Normal supply point / Material) it will return condition record number for condition type.
2. KONP (Conditions (Item)) .
3. J_1IMTCHID (Combination of Material Number and Chapter ID).
4. J_1IEXCTAX (Tax calc. - Excise tax rates
5 get the IR no. from RBKP and RSEG, and get the relevant FI doc no. and goto BSET table
for IR number to get from table apply logic for doc type RE ,you will get fi number nad in refernce field you will get invoice number + fiscal year
Map this by refernig on po for which grn and invoice happen
Regards
Kailas Ugale

Reports with 7000 records taking 10mts

Similar Messages

Maybe you are looking for