1 INFO-VAX	Sun, 02 Jul 2000	Volume 2000 : Issue 367       Contents:  Re: COE (was: OpenVMS loses big)  Re: COE (was: OpenVMS loses big)" Re: DECnet-Plus or DECnet Phase IV7 Re: Northern Light vs. Google (and the winner is . . .) 2 Re: OpenVMS loses big, was:  RE: Compaq advertises! Re: www.djesys.com is on the air!   F ----------------------------------------------------------------------  " Date: Sun, 2 Jul 2000 13:44:41 GMT0 From: "Terry C. Shannon" <shannon@world.std.com>) Subject: Re: COE (was: OpenVMS loses big) & Message-ID: <Fx2oo1.Kty@world.std.com>  L > In article <8jicv9$55r@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes:/ > > In article <K7lOF6yP86Ho@eisner.decus.org>, ; Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) writes: 7 > >>In article <88QiedE5WIdL@mccdev.slac.stanford.edu>, D Fairfield@SLAC.Stanford.EDU (Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515) writes: 1 > >>> In article <GQI1aC+b+oKV@eisner.decus.org>, ; Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) writes:  > J > >>>> Both Solaris and Tru64 are COE certified.  So is HP-UX.  Windows NT > >>>> is grandfathered in. ...  > >>>      ^^^^^^^^^^^^^^^^  > >>> J > >>>         Gack!  The new kid on the block is "grandfathered in"???  On > >>>     what basis?  > >>) > >>This decision was made in Washington. ( > >>Did you have any further questions ? > > G > > Just one - does the COE have any real value as a standard promoting K > > application portability?  Since Solaris sources are even less likely to J > > work out of the box on NT than they do on OpenVMS NOW, even before the COE A > > work, the value of this "standard" seems questionable indeed.   J The principal value of the standard IMHO is that it guarantees that CompaqL will support OpenVMS for a minimum of 15 years. The Q expects to garner overF $500M in incremental revenue from COE-ified OpenVMS over the next five1 years, but anyone can make revenue projections...    ------------------------------  # Date: Sun, 02 Jul 2000 14:04:12 GMT  From: jordan@my-deja.com) Subject: Re: COE (was: OpenVMS loses big) ) Message-ID: <8jni4m$pdc$1@nnrp1.deja.com>   , In article <8jm1l4$q0u@gap.cco.caltech.edu>,&   mathog@seqaxp.bio.caltech.edu wrote:4 > In article <395E66E5.787D745B@tmisnet.com>, Koloth <koloth@tmisnet.com> writes:D > >A lot of DOD and government contracts specify DII/COE compliance. HavingG > >OpenVMS Dii/COE certified overcomes one more obstacle to using it on B > >the contract.  Now if we can only get Compaq to market it...The other main obstacle  > D > That was exactly the same deal with POSIX, and then "poof" one day it'sF > gone and the only trace of its existence is the much despised "Open" in
 > OpenVMS. > G > At least while it was around Posix worked moderately well as you long  asH > you stayed inside that environment.  However, the one thing it did notB > address at all was the integration of OpenVMS machines into UnixB > environments.  That is, there was no NIS, and somewhat iffy NFS. > G > Will the COE work _finally_ add some reasonable way for a VMS machine  toF > become a NIS client (ie, login using the information it obtains from a NIS D > passwd map) and/or allow an OpenVMS machine to serve NIS maps?  If not,B > then it isn't going to do much to reduce the HUGE energy barrier against G > putting in an OpenVMS machine in a Unix environment, and that Everest  of aD > barrier is figuring out a way to make it play nicely with the Unix machinesF > in the same group.  I've got methods for keeping my VMS/Unix and WNTG > accounts in synch, but it's a _long_ way from plug and play, and I've  never F > been able to figure out any way to do things like "automount" on theE > OpenVMS end.  Conversely, a group with a bunch of Unix machines can  justC > buy another Unix machine, and know that making it available comes  down to G > not much more than telling the new machine where the NIS maps are and ) > telling the servers to let it see them.  > 7 > So can anybody at Compaq tell us if COE provides NIS?  > F > The COE at this point is all smoke and mirrors.  Any chance we could get E > some sort of a brief checklist of what it will have in it, and what  it > won't?  = I can't and won't speak to what Compaq plans to do to provide ; COE compliance for OpenVMS, but the standards documents for $ DII/COE compliance are available at:  !       http://diicoe.disa.mil/coe/   8 It's somewhat interesting to me that, I believe, DII/COE: compliance _requires_ that a large set of POSIX interfaces: be supported.  I don't know if this means the resurrection; of OpenVMS POSIX or some other way to meet this requirement 
 will be seen.   9 I do know that this is not _just_ checklist stuff.  There : is a testing and certification program which requires that5 a large body of configuration and management software , will run in a DII/COE compliant environment.  ' I don't know if NIS, etc. are included.    > 
 > Regards, >  > David Mathog > mathog@seqaxp.bio.caltech.edu @ > Manager, sequence analysis facility, biology division, Caltech >    -Jordan Henderson  jordan@greenapple.com     & Sent via Deja.com http://www.deja.com/ Before you buy.    ------------------------------   Date: 2 Jul 2000 09:05:45 -0500 9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) + Subject: Re: DECnet-Plus or DECnet Phase IV + Message-ID: <3dHWX9ZPt6J2@eisner.decus.org>   c In article <0d1001bfe3e2$dee35e20$020a0a0a@xile.realm>, "John E. Malmberg" <wb8tyw@qsl.net> writes:   K > It has more features than it's predecessor, and is backward compatable in N > communications with Phase IV nodes, and almost compatable in management.  ItH > has an ISO standard command language that is quite different to learn.  G But "different" is not the same as "difficult".  Most of the commentary E we have about learning the DECnet-Plus command language is from those % who knew the Phase IV language first.    ------------------------------  $ Date: Sun, 2 Jul 2000 00:23:47 -0700* From: "Nikita V. Belenki" <kit@nospam.net>@ Subject: Re: Northern Light vs. Google (and the winner is . . .)9 Message-ID: <b9C75.1037$0x.32275@nuq-read.news.verio.net>   2 "Bill Todd" <billtodd@foo.mv.com> wrote in message" news:8jmhop$r9i$1@pyrite.mv.net...  L > So:  They have something like (WAG) around 240 TB of total storage.  Split 3 I > ways (around their 3 sites) that's around 80 TB per site.  Mirrored (at  each; > site) gives them around 40 TB per site effective storage. G > Yes, that's still between 2 and 3 times what you estimated they might K > actually need, but it's getting close enough that one can start wondering H > whether your estimate could be off by somewhere near that much - or ifJ > Google simply thinks they *need* three- (or even four-) way mirroring toL > minimize the risk of losing data should two related disks (or one disk and a 4 > system handling a related disk) fail concurrently.  I Probably that's not the disk space, but the number of physical devices is D what they need to increase. *If* (no, I don't know their disk accessG statistics) there typically are several outstanding I/O requests to one H physical disk, their IDE disks are less effective than SCSI ones, so theH number of the physical devices should be kept higher to achieve the same sustainable access rate.  J Also they may have problems with the effective *cache* balancing, so theirF disk access rates may need to be higher than if their systems had less partitioned memory space.    Kit. kit # kits.net   ------------------------------  " Date: Sun, 2 Jul 2000 13:31:07 GMT0 From: "Terry C. Shannon" <shannon@world.std.com>; Subject: Re: OpenVMS loses big, was:  RE: Compaq advertises & Message-ID: <Fx2o1D.Fq5@world.std.com>  E > Oh.. you just want some of the utilities and other features?  Okay. . > Sector 7 does stuff like this, among others. >   6 Others including Accelr8 Technology at www.accelr8.com   ------------------------------   Date: 2 Jul 2000 09:00:12 -0500 9 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) * Subject: Re: www.djesys.com is on the air!+ Message-ID: <21p7mDeCCy8T@eisner.decus.org>   g In article <395E9EC8.4FBF49FC@earthlink.net>, "David J. Dachtera" <djesys.nospam@earthlink.net> writes:   & > http://www.djesys.com/vms/cdrom.html > I > The CD-ROM burning info. has been been re-organized and expanded a bit. H > I've added info. on CDRECORD and InfoServer "RECORD". However, I've noG > first-hand experience with either one. I took the CDRECORD info. from H > the Linux HOW-TO, and thr InfoServer info. is straight from a manual II > found on one of my current client sites. So, if those in the know would J > be so kind as to post some expanded/corrected info. here, I'll add it to: > the page with proper credit for the source. (Hi, Brian!)  G The InfoServer material looks correct to me.  As a confirmed InfoServer E user, the one quibble I would have is that the CD-R devices supported G are rare only in that they are no longer available _new_.  There should D be plenty of such devices available _used_ since they are as fast asG 2X.  In particular, I would bet total sales of Sony CDU926S units _far_ 3 exceeded total sales of InfoServers over the years.    ------------------------------   End of INFO-VAX 2000.367 ************************