[plug] dd speed estimates and /dev/hdx grepping

[Craig Ringer]

> | At the default of 512 bytes, each sector takes a number of reads and
> | writes to complete. Not terribly efficient. 
> 
> Really?  Craig said something along those lines too, but I'd've thought
> that the kernel would be clever enough to lump all the reads and writes
> together when it actually talks to the hardware.

AFAIK 2.4 is bad at efficient disk I/O in large chunks anyway - it 
splits up large operations and tries to re-merge them later, instead of 
retaining them all the way through the process. That said, while it 
might do that you're still looking at a lot more individual requests 
being queued and handled by the kernel, and it's generally messier. I 
don't know exactly why, but larger block sizes /will/ make your disk 
image operation more efficient, at least unless you have really funky 
hardware.

Craig Ringer


_______________________________________________
plug mailing list
plug at plug.linux.org.au
http://mail.plug.linux.org.au/cgi-bin/mailman/listinfo/plug