home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.arch:9400 comp.parallel:2115
- Newsgroups: comp.arch,comp.parallel
- Path: sparky!uunet!wupost!darwin.sura.net!gatech!hubcap!fpst
- From: dre@ember.Eng.Sun.COM (David R. Emberson)
- Subject: Re: sgi/sun shared memory machines question
- Message-ID: <lb23meINN7rg@appserv.Eng.Sun.COM>
- Followup-To: comp.arch,comp.parallel
- Summary: Sun MPs can run parallel codes.
- Sender: fpst@hubcap.clemson.edu (Steve Stevenson)
- Nntp-Posting-Host: ember
- Organization: Sun Microsystems Inc., Mountain View, CA, USA
- References: <1992Sep11.121830.17069@hubcap.clemson.edu>
- Distribution: usa
- Date: 11 Sep 1992 21:25:34 GMT
- Approved: parallel@hubcap.clemson.edu
- Lines: 23
-
- In article <1992Sep11.121830.17069@hubcap.clemson.edu>, zilla@nj.nec.com (John Lewis) writes:
- > Someone told me that the SGI multiprocessor machines cannot run
- > different unix processes on different processors; they can only
- > run a single (but possibly parallelized) process at a given time.
- > The reason was said to be that the machines do not implement
- > any mechanism for simultaneous writes.
- >
- > Is this correct? If so, how are simultaneous writes from
- > different threads of a parallelized process prevented?
- >
- > Is the new sun multiprocessor (4/690?) similar?
-
- This sounds bogus to me. All Sun MP machines (4/6xx, SS10, and futures) can
- run different processes simultaneously and also run parallelized codes. I
- think your friend is confusing "simultaneous writes" with synchronization.
- I believe the SGI machines do something funky to implement barriers (like
- put them in a special portion of memory with hardware support) because the
- R3000 has no synchronization primitives. This is not a problem on the Suns
- since Sparc has the hooks in the architecture. R4000 fixes this omission.
-
- Dave Emberson
- dre@Eng.Sun.COM
-
-