Hi everyone,
I have the sSS (0.4.5 from repo) running on a headless Ubuntu (14.04 Trusty) Server. It works well in general, when looking closer I have 2 issues:
1) When running, the ScannerServer generates a constant CPU load of about 5-10% (on an ASUS C60M1-I Mainboard /w dual core AMD C60 1GHz). Even if this mainboard is not exactly a screamer, this looks a bit much to me, considering that the scanner server does not do much until a scan job is triggered. Right? Does someone of you have an idea what might cause this?
2) I noticed, that while scanning from the ADF works, the next page is only pulled into the feeder after the last page was written to disk. This again takes quite a bit of time, which might be caused because the image transfer from the scanner to the server seems quite slow. This seems not to be a limitation by the scanner - scanning to an iPad is 5 times faster. (Scanning the same 5 pages (300dpi multipage pdf) took 155 sec to the linux server compared to 31 sec to the iPad). Might this have to do with how the image data is piped through? Would it be possible to eliminate 'wait states' in the program run?
Best
cmer
I have the sSS (0.4.5 from repo) running on a headless Ubuntu (14.04 Trusty) Server. It works well in general, when looking closer I have 2 issues:
1) When running, the ScannerServer generates a constant CPU load of about 5-10% (on an ASUS C60M1-I Mainboard /w dual core AMD C60 1GHz). Even if this mainboard is not exactly a screamer, this looks a bit much to me, considering that the scanner server does not do much until a scan job is triggered. Right? Does someone of you have an idea what might cause this?
2) I noticed, that while scanning from the ADF works, the next page is only pulled into the feeder after the last page was written to disk. This again takes quite a bit of time, which might be caused because the image transfer from the scanner to the server seems quite slow. This seems not to be a limitation by the scanner - scanning to an iPad is 5 times faster. (Scanning the same 5 pages (300dpi multipage pdf) took 155 sec to the linux server compared to 31 sec to the iPad). Might this have to do with how the image data is piped through? Would it be possible to eliminate 'wait states' in the program run?
Best
cmer