The biggest single factor that influences performance is the number of pixels that the image consists off. The larger the number of pixels, the more time it will take to process the image. Therefore, it is good to check your resample settings, if they are set correctly for your workflow. Read “To resample or not to resample“.
You can also set a limit on the size of images, so that individual images are not holding up your production. See screenshot below of those settings, which can be found in SIZE settings.
(MPixel = Mega-pixel or 1 million pixels).
System requirements and OS
Note that besides the obvious (processor speed and RAM), I/O performance is very important too. (speed of writing to and reading from hard-drive). SSD drives are great for performance. Especially if you work in virtual environments, make sure the I/O speed is good. The same goes for network speed if you use folders on the network.
Concurrent image processes
Pixometry can process up to 4 images simultaneously. The number of concurrent processes can be set during installation, or by editing the jClaro.conf file. Please note that it does not make sense to configure a number of concurrent processes that is higher than the number of processor-cores that are available. That will just slow things down. Also, make sure that there is sufficient RAM available in the machine to support the number of concurrent processes. 4GB per concurrent process, plus 4GB additional for the OS is a good guideline.
Configuring the number of concurrent processes and memory
To set these using the installer, read this: Installation. Note that you can run the installer again, and install over the existing installation without losing any of your configuration. So it is a valid way of changing those settings.
To change this by editing the jClaro.conf file:
This file can be found in the installation directory:
../Pixometry/conf/jClaro.conf
Open this file in your favourite text-editor. Locate these lines:
wrapper.java.additional.6=-Dcom.elpical.concurrent=4
AND this one:
wrapper.java.additional.15=-Xmx16384m
We advise to use 4096m memory per concurrent process. So in this case 4 * 4096m = 16384m.
Change the values and save the file. A restart of the Claro service is required for the changes to take effect.