In
<77142d37c0c3c34da0d7b1da7d7ca34308822...@nwt-s-mbx2.rocketsoftware.com>,
on 02/22/2011
at 04:06 PM, Bill Fairchild <bi...@mainstar.com> said:
>CONTIG and ROUND are/were independent, but they both addressed I/O
>performance improvement. One purpose of ROUND was to reduce DASD
>revolutions that did not transfer any data. CONTIG would provide the
>same result, but less often than ROUND, by reducing the number of
>times that the EOV service would have to be called to switch from one
>allocated extent to the next. The EOV macro results in an SVC,
>which, in the ancient days of OS/360 and SLEDs, might have involved
>loading the SVC code into the transient SVC area again (not sure
>which SVC type it was), which could result in an extra revolution on
>the user's data device.
Support for CKD I/O is a bit more complicated than that. EOV is a type
4 SVC, but it is not invoked just to switch extents. However, an
end-of-extent appendage is invoked, so there is some CPU saving in
minimizing extents.
ECKD, of course, significantly changed the performance tradeoffs. It
does not, however, eliminate the EOV needed to allocate a new extent.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT
ISO position; see <http://patriot.net/~shmuel/resume/brief.html>
We don't care. We don't have to care, we're Congress.
(S877: The Shut up and Eat Your spam act of 2003)
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@bama.ua.edu with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
