Anton wrote:
Let me spam too :)
the trick is to change buffer size. let's say bs=1024k will be much
faster.
And you do have dd for windows.
It is more efficient than count=$bytes bs=1 but not nearly as fast as
using count=0 seek=$bytes (i.e. creating a sparse file) as buffer size
doesn't effect performance when count=0 (it only acts as a multiplier
for the seek on the output file).
You only get a single ftruncate call when using the seek method instead
of read, write, read, write, ... in a loop.
It only takes a few milliseconds to create a 1TB file vs many minutes
for the other methods.
Although I must say this method's usefulness is limited if you want to
know you have the space actually available and allocated to the file
(but it is a very good method for testing large file support on your
file server / apache server etc).
~mc
On 26/07/07, Michael Clark <[EMAIL PROTECTED]> wrote:
Anand Vaidya wrote:
> Linux Way of doint it:
> dd if=/dev/zero of=filename.txt bs=1 count=$bytes
>
>
Couldn't resists optimising this further using sparse files :)
dd if=/dev/zero of=filename.txt bs=1 count=0 seek=$bytes
See how fast it is to make a 1TB file this way e.g.:
time dd if=/dev/zero of=filename.txt bs=1 count=0 seek=1T
~mc
_______________________________________________
Slugnet mailing list
[email protected]
http://www.lugs.org.sg/mailman/listinfo/slugnet
_______________________________________________
Slugnet mailing list
[email protected]
http://www.lugs.org.sg/mailman/listinfo/slugnet
