home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.unix.wizards:3864 comp.unix.programmer:4592 comp.unix.internals:1784
- Path: sparky!uunet!europa.asd.contel.com!darwin.sura.net!wupost!sdd.hp.com!swrinde!elroy.jpl.nasa.gov!ncar!noao!rstevens
- From: rstevens@noao.edu (W. Richard Stevens)
- Newsgroups: comp.unix.wizards,comp.unix.programmer,comp.unix.internals
- Subject: Re: Implementation of Sys V. based message queues
- Message-ID: <1992Sep9.195036.25030@noao.edu>
- Date: 9 Sep 92 19:50:36 GMT
- References: <1992Sep3.232807.18181@noao.edu> <1992Sep8.171014.21414@dickens.com> <18kvotINN7le@smurf.smurf.sub.org>
- Sender: news@noao.edu
- Organization: National Optical Astronomy Observatories, Tucson, AZ, USA
- Lines: 16
- Nntp-Posting-Host: gemini.tuc.noao.edu
-
- >Does anybody use message queues anyway? For something that can't be done
- >just as easily with named pipes or Unix-domain sockets?
-
- The only things you get with System V message queues are record
- boundaries (which you get with datagram Unix domain sockets,
- but then you have to provide your own flow control) and the ability
- to assign a type (e.g., priority) to each message, and use this
- priority to choose which record to read next.
-
- But, as others have already pointed out, there are *so many* drawbacks
- to using them that new applications should avoid them like the plague.
- Lots of people used to claim a need for them since they used to be
- fast, but today they are no faster than Unix domain stream sockets
- or SVR4 stream pipes.
-
- Rich Stevens (rstevens@noao.edu)
-