Hi.
Another idea would be to fork off some threads to do the splitting in
parallel. If the disk drivers and/or drives are brilliant, the reads
should get ordered such that you pay a minimum penalty for rotational
latency. The idea is that if there are two reads to be done on the
same cylinder, then the one rotationally closest to the head will go
first.
Dave
_______________________________________________
Bug-ddrescue mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-ddrescue
