Unlock a world of possibilities! Login now and discover the exclusive benefits awaiting you.
Hi All,
When I try to reload data from two different machines to same data source through oledb, the reload time varies a lot.
Below is time difference and I feel it because of the RAM size. Can anyone let me know if it is because of the RAM size the time is getting shorten or is there any other possiblitiy?
30min -> 16GB RAM
5 mins -> 128GB RAM
Thanks,
Rakul.
Hi!
Do you mean RAM on DB machines?
I think the answer depends on DBMS type.
DBMS has to fetch data to give it to qv.
Enough RAM allow DBMS keep most used data in memory and don't get data from hard drive.
The data being loaded from your database will populate a QV table in RAM.
If that QV table in RAM exceeds 16GB then on your 16GB RAM machine it will start swapping to the page file and swapping don't half slow QlikView down. But your 128GB RAM may well be handling it without swapping.
On the 128 GB RAM machine do the reload, and monitor it with with the Windows Task Manager and see how much RAM the qv.exe process max's at.
Hi.
I meant RAM of the computer that runs QV desktop. In my scenario, I m connecting to same DB using same connection type from two different machines and the time it takes is getting different. I m not sure if it has to do with RAM of the machine that runs QV desktop or is there any other factor that I m missing here.
Thanks,
Rakul.
Hi,
In either of the machine the RAM or CPU usage is overloading. But still the reload time is much smaller in bigger RAM machine. So need to know what are all factors that contribute to QV Reload from QV desktop using OLEDB connection type.
Thanks,
Rakul.
Does not matter whether or not it is OLEDB or not.
The more you overload your RAM the slower it will get.
is there the same (or similar) difference when you extract with a different tool (not Qlik)?
Can I read your statement as, Bigger the RAM lesser the Reload time?
Thanks,
Rakul
Not exactly.
If the server runs short of RAM and starts swapping then things will get slower.
But if the server has enough RAM then adding more will make no difference.
Hi,
You are using same connections. Are you using the same user credentials or different user?