On Sat, 2011-02-05 at 19:10 -0600, R. Ramesh wrote:
[...]
> So what does vmstat -d tell me? Is the number of IO under "total" column
> supposed to be the number of IOs
> issued to the controller with each IO being contiguous N sectors?
"man vmstat" says 'total' is "Total writes completed successfully".
Vmstat shows the same numbers as "cat /sys/block/sda/stat" (presumably
where it gets them from) and a copy of the linux documentation for this
is at http://www.mjmwired.net/kernel/Documentation/block/stat.txt
So yes, it looks like 'total' is the number of contiguous blocks of read
or writes. (Though they may not be contiguous once they get onto NAND.)
And 'sectors' the total size of all data read or written.
Thanks. I will have to assume that each contiguous write does not cause
more than two erase-block writes since I do not believe single kernel
write is more than the size of erase block. Any way, it is either 24
years or 240 years. That is, either I will be very old or my great grand
children will be very old when this happens. In other words, it is time
for me to stop worrying, like you said :-)
Thanks for helping. It was an interesting exercise for me. Hopefully,
you enjoyed it too.
Ramesh
--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/4d4ec6f1.6070...@ti.com