home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.parallel
- Path: sparky!uunet!think.com!rpi!gatech!hubcap!fpst
- From: jf@edv6000.tuwien.ac.at (Josef Fritscher)
- Subject: Re: PVM vs. Express
- Message-ID: <1992Nov16.131843.5549@hubcap.clemson.edu>
- Apparently-To: comp-parallel@news.uu.net
- Sender: news@email.tuwien.ac.at
- Nntp-Posting-Host: edv6000.tuwien.ac.at
- Reply-To: fritscher@email.tuwien.ac.at
- Organization: Technical University of Vienna
- References: <1992Nov13.125101.20219@hubcap.clemson.edu>
- Date: Mon, 16 Nov 1992 11:17:08 GMT
- Approved: parallel@hubcap.clemson.edu
- Lines: 52
-
- In article <1992Nov13.125101.20219@hubcap.clemson.edu>, Roy_Williams@nntp-server.caltech.edu writes:
-
- |> Portability: PVM seems to be just workstation networks, Express is also Intel, Ncube
- |> Cray, SGI, in fact all the MIMD machines I can think of.
-
- On the other side you can mix workstations from different vendors
- (heterogeneity) with PVM but not with Express.
-
- |>
- |> Debuggers: Express has ndb, PVM has the usual unix debuggers. You have to learn
- |> ndb, but its built as a parallel debugger (Firing up 64 dbx's could be a bit of
- |> a strain on any machine I would think)
-
- The new (public domain) tool xab will give you debugging and performance
- monitoring features.
-
- |>
- |> Express has a reasonable I/O model: the processors contribute to the file in order,
- |> or all the processors speak with one voice, or they all write asynchronously, and
- |> similarly for reading. I've found it really awkward trying to write an intelligible
- |> file when anyone can write whenever they want.
-
- This only needs some synchronisation stuff which can be written in PVM too.
-
- |>
- |> PVM seems to be just send and receive and not much else. You can't send a "mixed-bag"
- |> message with different types - you have to send the floats, ints, chars etc in
- |> different messages. Also there seems to be a kind of hub-and-spoke system where all
- |> messages get routed through the hub (is this true?).
-
- Completely wrong. You can mix of different types of data in one message and you
- have barriers, signals, etc.. Also the routing is no problem because PVM is
- based on TCP/IP where no routing is neccessary. Of course messages are
- transported through a daemon, but that is in most cases an advantage.
-
- BTW. the new version of PVM (3.0) will address most of the features addressed
- above.
-
- - Joe
-
- --
- Josef Fritscher Internet: fritscher@edvz.tuwien.ac.at
- Computing Center Bitnet: fritscher@awituw64.bitnet
- Technical University of Vienna in reality: jf@edv6000.tuwien.ac.at
- Wiedner Hauptstrasse 8-10 Voice: ++431 58 801/5505
- A-1040 Vienna, Austria/Europe Fax: ++431 58 74 211
- Member of the Austrian Center for Parallel Computation (ACPC)
- -------------------------------------------------------------------------
- The function of local-in-Universe critical information-gathering and
- local-in-Universe problem-solving is manifest in the cockpit of all
- great airplanes. (R. Buckminster Fuller)
-
-