Find Answers
Qlik Community
Collaborate with over 60,000 Qlik technologists and members around the world to get answers to your questions, and maximize success.
Join Us
Knowledge
Task reload on the QlikView server/ Publisher Server are slower than on the QlikView desktop
Article Number: 000055228 | Last Modified: 2018/11/21Description
Re-loading of a document takes longer on a Server machine than on a PC running Desktop.
Re-loading is slower running a QlikView Desktop on the QlikView server itself also
Enviornment:
QlikView all versions
Resolution
It is not unexpected for a reload in a QlikView Publisher to take a bit longer than doing a reload in QlikView desktop.
This is because the publisher does more than just the reload. It must open the document and also potentially distribute it. All of these operations combine to the overall task duration time.
You can open the task log and see how long the actual reload takes.
But if the reload takes much longer on a QlikView Desktop on the server machine also, it could be because of the hardware configuration on the machines.
Server machines often have more than one CPU, with many CPU cores but they might have a lower clock frequency than the CPU cores on a PC used for QlikView desktop.
Some of the operations executed in a reload script are single-threaded, meaning that only 1 of the CPU cores in the server is used. For this reason, clock frequency will have a heavy impact on execution time.
So for example:
The Machine that is running the QlikView Desktop reload is a 1 CPU 2 core 4 logical processors machine with 3.0GHz clock frequency.
The QlikView Server has 4 CPUs, each with 8 Core 16 logical processors with 2.4GHz clock frequency.
When you do a reload and you run in to a single-threaded operation, only 1 of the 64 processors get used for that operation.
Since the Desktop has faster CPU it does this faster than the server.
Also in the example above the machine is using hyper-threading, this is normally not recommended for 4-socket machines. For recommendations on hyper-threading see
BIOS Settings For Best Performance On QlikView Server And QlikView Publisher
For more information on some single threaded operations see;
Single/Multi-threading in QlikView
Sum() with "GROUP BY" is single-threaded calculation
This is because the publisher does more than just the reload. It must open the document and also potentially distribute it. All of these operations combine to the overall task duration time.
You can open the task log and see how long the actual reload takes.
But if the reload takes much longer on a QlikView Desktop on the server machine also, it could be because of the hardware configuration on the machines.
Server machines often have more than one CPU, with many CPU cores but they might have a lower clock frequency than the CPU cores on a PC used for QlikView desktop.
Some of the operations executed in a reload script are single-threaded, meaning that only 1 of the CPU cores in the server is used. For this reason, clock frequency will have a heavy impact on execution time.
So for example:
The Machine that is running the QlikView Desktop reload is a 1 CPU 2 core 4 logical processors machine with 3.0GHz clock frequency.
The QlikView Server has 4 CPUs, each with 8 Core 16 logical processors with 2.4GHz clock frequency.
When you do a reload and you run in to a single-threaded operation, only 1 of the 64 processors get used for that operation.
Since the Desktop has faster CPU it does this faster than the server.
Also in the example above the machine is using hyper-threading, this is normally not recommended for 4-socket machines. For recommendations on hyper-threading see
BIOS Settings For Best Performance On QlikView Server And QlikView Publisher
For more information on some single threaded operations see;
Single/Multi-threading in QlikView
Sum() with "GROUP BY" is single-threaded calculation
Get Answers
Find Answers
Get Support
Have a Question?
Search Qlik's Support Knowledge database or request assisted support for highly complex issues.
Submit a case
Talk with a Representative
Critical Issues
Experiencing a serious issue, please contact us by phone. For Data Integration related issues please refer to your onboarding documentation for current phone number.
Call Us