On Tuesday 10 January 2006 21:05, Rony Bill wrote:
The application is as follows. Taking a video clip from a camera as well as a VTR,
This is the capture part and is very io intensive. This will happen on a standalone machine. 25 fps D1 10 bits per color. 25*720*576*3*2=62.2MBps. Regular pci bus will choke. U require hardware compression on the capture card or PCIE. And fast storage 230 GB per hr of raw video.
processing it in the machine
This can be distributed on the cluster and will scale linearly.
and editing it.
Same here. Both the above depends on what processing u are doing. Basically each frame is divided into blocks. Figures are available on the net for compute cycles per block and per frame. U can distribute blocks or frames across the cluster. There are tradeoffs.
After edit, its converted to dvd for backup and broadcast. One second of uncompressed video is 50 MB in digital format.
See above.
From all this data, how does one calculate the above 3 parameters.
Regards,
Rony.
To help you stay safe and secure online, we've developed the all new Yahoo! Security Centre. http://uk.security.yahoo.com