Stan,
> Using MR/NOBUF (18 writes):
> plain CPU = 180, elapsed 207 milliseconds.
> XM CPU = 405, elapsed 2,022 milliseconds.
> fcontrol 2 CPU = 168, elapsed 765 milliseconds. (18 calls)
>
> Using ordinary record writes (14,033 writes):
> plain CPU = 1,395, elapsed 1,440 milliseconds.
> XM CPU = 9,246, elapsed 10,471 milliseconds.
> fcontrol 2 CPU = 9,913, elapsed 118,472 milliseconds.
The difference between those times are impressive. What's the
difference between using XM and fcontrol mode 2 regarding data
protection during a system or program failure?
- Cortlandt
"Stan Sieler" <[log in to unmask]> wrote in message
news:39b82340$1_2@skycache-news.fidnet.com...
> Re:
>
> > So, are these files covered by XM? Apparently not, though it is
difficult to
> > find doc on the XM.
>
> The file label is covered, not the file data ... unless you manually
attach
> it to the "serial write queue" via fsetmode.
>
> > If we really want the data to be in the files, do we end up having
to do a
> > fcontrol mode 2 to set complete IO? Or, this there a better way to
make sure
> > we have the data in the files if the system happens to fail? I
would think a
>
> Time to copy a 1 MB file (using MR NOBUF, taking about 56
reads/writes):
>
> Using MR/NOBUF (18 writes):
> plain CPU = 180, elapsed 207 milliseconds.
> XM CPU = 405, elapsed 2,022 milliseconds.
> fcontrol 2 CPU = 168, elapsed 765 milliseconds. (18 calls)
>
> Using ordinary record writes (14,033 writes):
> plain CPU = 1,395, elapsed 1,440 milliseconds.
> XM CPU = 9,246, elapsed 10,471 milliseconds.
> fcontrol 2 CPU = 9,913, elapsed 118,472 milliseconds.
>
> Note: this was on a single user 968 system with 256 MB of memory,
> running MPE/iX 6.0
>
> Stan Sieler
[log in to unmask]
> www.allegro.com/sieler/wanted/index.html
www.allegro.com/sieler
>
|