BPC 10 - EPM data retrieval very slow!

Hi BPCers,
We are using an Excel EPM Input Schedules as a Resource Management tool - using VBA to provide the functionality we need.
Performance is generally good, but quickly deteriorates when handling larger data sets - even 500-600 rows of transactional data is enough to slow data retrieval from our BPC Cube to EPM to an unusable speed. This is in relative terms pretty small so there should be some option for optimisation.
Does anybody have any experience with this? All suggestions welcome. We are operating on EPM Service Pack 7 Patch 1, but I'm not sure that EPM is necessarily the problem here.
Thanks,
Tom

Thanks Gersh,
Had a look through fiddler and have identified the job that is causing the delay - some rooting around in the ABAP debugger produced the answer as to why adding more data slows processing speed so dramatically.
When we take data from the back end, we select a couple of parameters which limit the range of data that we are pulling through - a certain set of people, and a certain range of days. Once this is pulled through, allocations are made to any combination of person and day within this range, which generates and extra two properties - a project ID and a work status.
This makes 4 properties, and when BPC pulls data it attempts to find every combination of every one of the properties that exist within this range - so the more allocations made the more this slows down as it dramatically increases the number of combinations.
The result is that BPC runs through a couple of hundred thousand generated tables, most of which are nonsense.
Not sure what to do from here. This is how BPC reads data so approaching a fix could be difficult.
Tom

Similar Messages

Fetching data is very slow in workspace for planning application

Hello Everyone,
I am working in the web environment of a planning application and recently I have seen some issue's like Fetching data is very slow through webform's and the system will locking automatically and which kick me off from workspace when I refresh.So could you suggest me like what would be the reason's for this issue's.
Thanks in advance

Hi,
Sounds like your form is very large or there are maybe a lot of dynamic calcs on the retrieval. How many rows and columns have you on the form ? Have you dense or sparse dimensions as rows or columns?
Brian

Data transfer very slow on mac pro!! mac H/W unable to fit specifications.

Hi,
I am installinginternal disks on the quad mac pro and the transfer ratet in the mac are very slow!!!
the disks are: barracuda 7200.10 disks, 500 GB. Sta II 300, max ST3500630AS firmware 3.AAE . throughput:300mb.jumpred off to get the 300.
MAC tells that the QUAD MAC Pro equipment (2x2.66ghz dualcore intel xeon) comes with a 3.Gigabit specification/ 300MB/sec. (Intel ESB2 AHCI, speed: 3.0 Gigabit. AHCI version 1.10 supported)...
And the benchmarks of the disk ST3500630AS show maximum 80MB/sec and a minimin of 37.3MB/sec, average 63.1 MB/sec. (anandtech tests).
According to seagate.com the maximum sustained data transfer rate:78MB/s
but in the mac the reallity is different: I am getting a transfer rate of 22MB/sec!!!!! . with out any application running and only making copy of the files!!!!
that means that the mac is only providing only 1/3 of the expected transfer rate for the disks..
so, is there a way that the MAC Pro gets all the normal transfer rate for those disks???
Is the Quad MAC Pro hardware unable to get the maximum throughput of the disks???
It seems that the mac is unable to get also the average transfer rate of those disks.. how to correct it???
thanks a lot
Alberto
ps. just for fun: I am getting a faster transfer rate from an old power pc ,and old pc, and firewire 400 and old ide disks!!! (36MB/seg!!!)

Yup.
BThe latest Mac Pro are better, so if you have a new or 2 month old unit, Seagate should work, though I think there should be later firmware.
Some would say those drives take the cake or woRSe.
http://www.barefeats.com/hard91.html
SEAGATE PUZZLES
+Firmware version 3.AEE or later solves the slow sustained large block write speed issue for a single 7200.10 inside the Mac Pro.+
+The remaining performance issue is slow small random read speeds for one, two, three or four drives. No matter how many drives you configure in RAID 0 sets, the average random read speed for combined block sizes from 64K to 1024K is less than 30MB/s (based on QuickBench 3 testing).+
+Until Seagate fixes this, we can't recommend the 7200.10 series as the ideal boot drive for the Mac Pro (or Power Mac).+
http://www.barefeats.com/quad08.html

Data extraction very slow

Hello All,
I am trying to extract data from init 2lis_11_vaitm. But, extraction is very very slow. Total data in init set-up table are only 200k.
One thing I noticed is that we changed the extractor to increase a no. of fields to 225. All these fields were available in the BI content delivered communication extract structure of 2lis_11_vaitm in ECC.
Is it because of large nos of fields being extracted from ECC. 200k records are being extracted..It has been 38 hrs..and so far only 140K records are extracted.
Could you please suggest, how I can improve extraction performance.
We will be moving this to production. I am afraid that much larger nos of records in production will take for ever to extract.
Thanks
Shailini

Yes you are right, IO = input and output, and generally reference to the data loading capability
BASIS will help to monitor the data loading capability.
They will help to check the sizing document, prepared before system go launch, to check the "designed capability". BASIS guys need to do some test on data loading before system go launch.
And they will help to check the log to find some problems during data loading.
In bi 7 statistics you can find some information about that load. Discuss with the basis guys, they can help to analysis the problem without the BI statistics.
The system on hand do no load from 2lis_12_vcitm. But here is some information for your reference:
1. Production Server: 26K records, load to 2LIS_03_BF, takes 58s in all
2. Testing Server: 1.5 Million records, full load to 0FI_GL_10, Runtime 34m 12s

Large SGA issue-- insert data is very slow--Who can help me?

I set the max_sga_size to 10G,and db_cache_size to 8G, but the value of db_cache_size is negative number in OEM, and also I found that the data inserting was very slow, I checked the OS, found no CPU consuming and no IO consuming.
The OS is HP-UX B11.23 ia64.
Oracle server 9.2.0.7
Physical memory : 64G
CPU: 8
(oracle server and os are all 64-bit).
If I decrease the SGA to 3G,and db_cache_size to 2G, and the same data inserting is very fast, everything is well.
so I guess if there are some os parameters needed to set for using LARGE memory.
Who know this issue or who has the experience of using large SGA in HP-UX ?
Message was edited by:
user548543

Sounds like you might have a configuration issue on the o/s side.
Check that kernel parameters are set as recommended in the installation guide
The first thing that came to mind after reading the problem description was that you might have too low SHMMAX for that 10GB SGA, which would cause multiple shm segments to be created and thus explain the performance degration you're experiencing.
A quick way to check if that's the case would be doing "ipcs -m" to see if there are multiple shm segments when SGA is set to 10GB.

My data base very slow

Dear all,
I have oracle rac db time to time its going very slow i can't recognize why this is happen how to check this . When the db load is going high then application also restarted .
An that time load average going very high ,only one node

user11876003 wrote:
Dear all,
I have oracle rac db time to time its going very slow i can't recognize why this is happen how to check this . When the db load is going high then application also restarted .
An that time load average going very high ,only one nodeIts actually nothing that you have told us to suggest something. What's the db version and o/s? Oracle database won't restart any application if being slow. That you should take up with the application development team of yours. For the database, depending on your db version and license, get either a Statspack or AWR of not longer than 30 minutes and post here .
Aman....

Data is very slow

While I do love the phone, data is definitely a lot slower.

Viphone26 wrote:
Well compared to a Droid 2, a droid X, and an LG ally that my cousin has, data is quite slow. I find it taking a good couple of seconds to open up a web page, and it takes a little longer than usual to download an app via 3g than my ipod touch can via wifi.
Its probably their way to get you to justify buying another 700 dollar phone when the Iphone 4G comes out and data prices go through the roof.

Retrieval very Slow when using Partition in ASO

Hi - I have partitioned my ASO cube using Transparent partition and I amretrieving it thru BSO.The retrieval for some combinations are VERYVERY SLOW in Excel.We have over 250,000 members.I have usedRetrieval Buffer and Retrieval Sort Buffer sizes of 10000KB and1000KB respectively.Some retrievals take more than 4 minutes. When Designing Aggregation for ASO, I have used the Recommendedviews to materialize aggregation. Do I have to do or change some settings for better Performance? Any help would be greatly appreciated. Thanks, Narain

Making direct I/O (in place of default buffered I/O) candefinitely help a lot.Then you can increase some data file cache upto 2GB [as per yoursys support]. Also check about your dedicated parallel processsides.I wish it helps.

Data is very slow in my phone

Hi, I am using $45 montly go phone plan which has high speed data included and my device is Samsung Galaxy grand quattro. The data speed is very low. Most of the time I couldn't check my e-mails too. Is there any problem with the settings or device ?

Hello @bgvel123,
Sorry to hear you’re experiencing slow data speeds, I’d be glad to help you out with this.
Please visit our Troubleshoot & Resolve page, select your device make and model and choose the option saying Internet.
Follow the steps and you’ll be able to see what could be affecting your data speeds.
Thank you,
Charise

Data extraction very slow thru DB Connect

Hello all,
I am using DB connect to extract data from (200k/day) the Oracle source system. I have to extract 6months of historical data. Could any one please suggest me extraction parameters I should apply to improve extraction process.
Currently, DB extract is giving very poor performance. Moreover, it is ending in green w/o extracting all data. The key field in the source table is only a date field, which I am using in selection criterion.
Points will be assigned.
Thanks

You could create an index on date field for the table.

EPM on solaris very slow and stops responding

Hi,
We installed EPM 11.1.2.2 using a single Managed server (EPMServer0) on Solaris with 32Gig memory and use the start.sh script to start all components.
The applications runs fine initially but after second day it starts to slow down and then hangs. Even couldn't login after that. So we restart EPM and things back to acceptable performance.
We did install OHS on top WebLogic, with EssBase, Planning, Reporting and Analysis, etc.
In Solaris we notice the memory using by EPM grows to around 6Gig.
Didn't modify an weblogic heap size. Noticed when issue ps -ef two java processes:
/u01/oracle/Middleware/jdk160_29/bin/sparcv9/java -server -Xms256m -Xmx512m
/u01/oracle/Middleware/EPMSystem11R1/../jdk160_29/bin/java -Xms128m -Xmx768m
Tried to trace how start.sh starts the processes and it might using the following script to set the java options:
./user_projects/epmsystem1/bin/deploymentScripts/setCustomParamsEPMServer.sh
This has -Xms128m -Xmx8160m -d64 -XX:PermSize=64m (and many others)
Sure there are also setDomainEnv.sh and commEnv.sh but I don't see they are used with the start.sh script.
Appreciate suggestions on how to resolve this critical performance issue.
Thanks

Thank you. Everything is back to speed, except for backing up a catalog to a DVD. That's still extremely slow and then it stops. It also doesn't let me sign in even with a new password. Could Elements be corrupted? If it is can I uninstall it and reinstall it, and if I do will I lose the photos in my catalogs?
Another thought. Since everything else seems to be OK, could the problem be with my disc drive?

Data guard-- very slow replication.

Hi Experts,
I am facing problem in an Oracle 10G R1 data guard implementation running on Solaris 10 64bit platform.
The primary database site has got 2 Sun servers running on RAC while the parallel standby site has got exactly the
same hardware. A LOGICAL standby policy is in use.
->>>The standby database is taking some 30 mintutes to get 10-15 new insertions from the primary database, ironically the system is not under production so no question about systemload.
* How can I tune the data replication time gap between the primary site and the standby site.
* Any pointer to some documentation 'd be highly appreciated.
Many thanks
Nitin

Hi NItin,
Here is my Logical standby: SQL Apply too slow on this matter.
Did you make any changes to following settings of logical standby parameters:
EAGERSIZE
PARALLEL_MAX_SERVERS
APPLY_SERVERS
LCR Cache size i.e. MAX_SGA
Please read before Note:603361.1 and Note:312434.1.
Please let me know if you decide to change something. I can advise some more or less appropriate values but you need to test them thoroughly.
Best Regards,
Alex

Awm mantain data is very slow

HI,
I'm using AWM 10.1.0.4 on 10.1.0.4 db with OLAP option. When I try to maintan a cube of 4 dimensions and 25k data, it takes a lot of time (more then 8 hours). Is this normal? Have I to set some parameters?
CPU: Intel celeron 2,2Ghz
RAM: 1GB
Thank you in advance!

The issue here might be the time taken to aggregate your data. Within most multi-dimensional models information at the lowest level tends to be very sparse, for example not every customer buys every product every day. Oracle OLAP has some very powerful features to manage sparsity. In AWM as part of the cube definition process you can manage sparsity using the "Implementation Details" tab. I would recommend setting each dimension as sparse except your Time dimension. Time is typically not sparse in most multi-dimensional models, although there are exceptions to this rule.
If you only have 25,000 rows to load I would not recommend using compressed cubes or partitioning the cube over time.
You can control the aggregation step of the build process by selecting/de-selecting dimension levels in the "summarize To" tab. If you want to determine how much time is specnt build the multi-dimensional structures then do not summarize any levels above the leaf nodes. Then, summarize all levels and compare the two times. This will tell you how much time is spent actually creating and managing all the summary information.
Other recommendations to consider are to use surrogate keys when building your dimensions as this ensures uniqueness across all levels within your dimension and helps to avoid the creation of circular references in your parentage table.
Hope this helps
Keith
Oracle Business Intelligence Product Management
BI on Oracle: http://www.oracle.com/bi/
BI on OTN: http://www.oracle.com/technology/products/bi/
BI Beans http://www.oracle.com/technology/products/bib/index.html
Discoverer: http://www.oracle.com/technology/products/discoverer/
BI Software: http://www.oracle.com/technology/software/products/ias/devuse.html
Documentation: http://www.oracle.com/technology/documentation/appserver1012.html
BI Samples: http://www.oracle.com/technology/products/bi/samples/
Blog: http://oraclebi.blogspot.com/

I have 4 Bars, LTE is connected, But iPhone no data or VERY slow data

Hi, I have an iphone 5s. I have chatted and reached out to verizon about 8 times because weirdly my phone will have signal, show its connected to LTE. but data won't work. If i turn LTE off- the phone works fine. I've reset the phone, got a replacement sim card and am having the same issue.
today, data is working but its so slow... It's slower than 3G. I actually tested this by loading a youtube video LOL. The weird part is my coworker has a samsung galaxy s4... and is not having the same issue. IS this a known issue with iPhones?
I have a replacement phone on the way; but what could cause this? I have had this phone for just under 2 months and havent had any accidents or dropped it.

I came up on your post, I too am experiencing the same thing. I got the new iPhone 5s on SAT and today I started having no connection issues but the phone would say I have 3 to 4 bars of signal. I ran speed test one minute I would have 67 ping and 12MB download then it would drop to .02MB and 100 ping in the same spot in the room. I can restart the device and it will run well for about 2 min then it craps out. I have the unlimited data package and my droid phone never had this problem Im thinking of taking this thing back.

EPM Web Pages very slow to render

I've installed Foundation Services on a Windows 2008 box with a single dual core CPU, 4GB RAM.
When I launch Workspace, the web page takes at lease 5 minutes to render. In addition launching any applciation within Workspace e.g. Shared Services, it also take about 5 minutes to render.
Any suggestions on what I should check to resolve this?

You should only have one router on a home network (as its function is to translate a single public IP address to multiple private addressed on your LAN). If you have the 2Wire and TC both configured as routers, slow connection or no connection would be the likely result.
Suggestion would be to choose "Add New Network Device" when setting up the TC and add it as an Access Point. This will provide wireless connectivity and ensure that only one device is doing network address translation. My network is setup in this exact way (when an Airport Extreme router and a TC added as an Access Point) and I have not experienced any problems with 4 Macs that connect wirelessly.

BPC 10 - EPM data retrieval very slow!

Similar Messages

Maybe you are looking for