0 INFO-VAX	Mon, 15 Jan 2001	Volume 2001 : Issue 30      Contents:# Re: ACP_DATACHECK, was: Faster gzip # Re: ACP_DATACHECK, was: Faster gzip # Re: ACP_DATACHECK, was: Faster gzip ( Re: BJ's Warehouse club - using OpenVMS?( Re: BJ's Warehouse club - using OpenVMS?( Re: BJ's Warehouse club - using OpenVMS? BMC PAtrol still used? Re: BMC PAtrol still used? Re: BMC PAtrol still used? Re: CHARON-VAX Re: CHARON-VAX" RE: Cluster Alias under Decnet OSI4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 RE: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution4 Re: Compaq: A simple, affordable clustering solution Re: Dead VAXstation 3100 Re: dec 3000 300 crashing  Re: dec 3000 300 crashing / Re: DECNet dyslexia (was: Re: Version dyslexia) * Re: Faster gzip, thank you Paul Repacholi!* FILESERV@WKU: UnZip V5.42 is now available Re: GS160 hardware question  Re: GS160 hardware question  RE: GS160 hardware question  Re: GS160 hardware question  Re: GS160 hardware question  Re: GS160 hardware question  Re: GS160 hardware question  Re: GS160 hardware question  HIghest file version Re: HIghest file version Re: HIghest file version) HW collection (was Re: Dec 3000 300 help) / License - Dec C for OpenVMS AXP and broken link 3 Re: License - Dec C for OpenVMS AXP and broken link  Mhz of a Alpha CPU?  Re: Mhz of a Alpha CPU?  Re: Mhz of a Alpha CPU? ( Mistakes were made: was RE: Compaq: etc.9 Re: OpenGL on a Compaq XP1000 w/ Elsa Gloria Synergy card < Re: OpenVMS and DII COE (was Re: IBM + Linux + US$ 1 bilion)& OpenVMS Volume Management (On the fly)* Re: OpenVMS Volume Management (On the fly)* Re: OpenVMS Volume Management (On the fly)* Re: SYSGEN details missing from Audit logs Re: UCX Routing Problem  Re: UCX Routing Problem  Re: Variables in DCL Re: Variables in DCL3 Re: Virus scanning on VMS (was Re: OpenVMS on CNET)  vms tool Re: vms tool Re: vms tool Re: vms tool Re: vms tool Re: vms tool) What is the installed base of Tru64 Unix? - Re: What is the installed base of Tru64 Unix? 5 Re: what is the latest version for netscape on VMS... $ where is the new version of OpenVMS?( Re: where is the new version of OpenVMS?( Re: where is the new version of OpenVMS?( Re: where is the new version of OpenVMS?( RE: where is the new version of OpenVMS?( Re: where is the new version of OpenVMS?( Re: where is the new version of OpenVMS?G Re: [Change subject] percentage of world's processing on "mainframes" ?   F ----------------------------------------------------------------------   Date: 15 Jan 2001 16:40:28 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog), Subject: Re: ACP_DATACHECK, was: Faster gzip, Message-ID: <93v95s$s3b@gap.cco.caltech.edu>  _ In article <87g0iocb1q.fsf@k9.prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes: 5 >mathog@seqaxp.bio.caltech.edu (David Mathog) writes:  > E >> FASTRMS is a big help, and changing ACP_DATACHECK is a minor help  I >> (if your software is creating a lot of files), but the fundamental RMS > >> factor of 2 or 3 performance lag of RMS vs. Linux remains.  > D >Does the Linux code do a SYNC to get the data on the drive? If not,9 >using NLA0: is the equivalent for GARENTEED performance.  >   L The 2-3X was comparing Linux file cache speed with VMS RAMdisk speed.  That J is, memory to memory on both platforms through the file IO interface.  VMSJ physical disk speeds are yet another order of magnitude slower.   The 2-3XD slower performance is the best you can do with a simple C program onG OpenVMS that does nothing more than shuffle data to and from files with G minimal processing on it.  (Think "tr" or "head".) That 2-3X represents H some sort of overhead (maybe record locking, nested subroutine calls, orC multiple buffer copies) that is present in RMS and absent in Linux.   K The thing is that there are a lot of times when getting the data to disk is D not only unimportant, it isn't even desirable!  On Linux I can do anB operation and then sync, and know the data is where I want it, butJ currently on VMS I have no choice but to wait for the data to hit the diskK whether that's important or not. For instance, I do a lot of data crunching J on Linux that involves slinging around several >100Mb files, and it is notD important that even a single one of those files ever hits the oxide,K because I yank the final results file off the system and send it elsewhere, J and all the other files are just temporaries.  Now I could (if I had a tonG of money) put VMS on those DS10s, but I would not do so, because of the I 10-100X slowdown I'd suffer for these applications.   And even if I could H use most memory as RAMdisk (which is a lot less flexible than the Linux D file caching) there would still be at least a 2-3X slowdown.  If VMSK engineering would like to do something to make a lot of VMS customers happy K they'd find some way to squeeze some of this overhead out of RMS. I suspect G most VMS customers would be very pleasantly surprised by the subsequent K speed increases.  (The cynic might argue though that Compaq might not be so F pleased, since fewer customers would be shelling out the big bucks for3 heavy hardware solutions to their IO bottlenecks.)     Regards,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech     ------------------------------    Date: 15 Jan 2001 12:01:00 -0500* From: young_r@eisner.decus.org (Rob Young), Subject: Re: ACP_DATACHECK, was: Faster gzip+ Message-ID: <BAGKF9FTDVUJ@eisner.decus.org>   a In article <93v95s$s3b@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes: a > In article <87g0iocb1q.fsf@k9.prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes: 6 >>mathog@seqaxp.bio.caltech.edu (David Mathog) writes: >>F >>> FASTRMS is a big help, and changing ACP_DATACHECK is a minor help J >>> (if your software is creating a lot of files), but the fundamental RMS? >>> factor of 2 or 3 performance lag of RMS vs. Linux remains.   >>E >>Does the Linux code do a SYNC to get the data on the drive? If not, : >>using NLA0: is the equivalent for GARENTEED performance. >> > G > The 2-3X was comparing Linux file cache speed with VMS RAMdisk speed.   & 	What version of the RAMDISK software?   				Rob    ------------------------------   Date: 15 Jan 2001 18:44:21 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog), Subject: Re: ACP_DATACHECK, was: Faster gzip, Message-ID: <93vge5$4ls@gap.cco.caltech.edu>  X In article <BAGKF9FTDVUJ@eisner.decus.org>, young_r@eisner.decus.org (Rob Young) writes:b >In article <93v95s$s3b@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes:b >> In article <87g0iocb1q.fsf@k9.prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:7 >>>mathog@seqaxp.bio.caltech.edu (David Mathog) writes:  >>> G >>>> FASTRMS is a big help, and changing ACP_DATACHECK is a minor help  K >>>> (if your software is creating a lot of files), but the fundamental RMS @ >>>> factor of 2 or 3 performance lag of RMS vs. Linux remains.  >>> F >>>Does the Linux code do a SYNC to get the data on the drive? If not,; >>>using NLA0: is the equivalent for GARENTEED performance.  >>>  >>  H >> The 2-3X was comparing Linux file cache speed with VMS RAMdisk speed. > ' >	What version of the RAMDISK software?  >   
 DECRAM 2.3   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech     ------------------------------   Date: 15 Jan 2001 15:25:38 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) 1 Subject: Re: BJ's Warehouse club - using OpenVMS? , Message-ID: <93v4pi$1v99$3@info.cs.uofs.edu>  ; One of my favorite local Pizza joints is still using VMS on < HDS terminals for all it's POS operations.  I would bet that< they run their inventory control, payroll and AR/AP on it as well.    bill   --  J Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>       ------------------------------    Date: 15 Jan 2001 11:14:18 -0500/ From: jordan@lisa.gemair.com (Jordan Henderson) 1 Subject: Re: BJ's Warehouse club - using OpenVMS? * Message-ID: <93v7kq$gfm$1@lisa.gemair.com>  , In article <93v4pi$1v99$3@info.cs.uofs.edu>,- Bill Gunshannon <bill@cs.scranton.edu> wrote: < >One of my favorite local Pizza joints is still using VMS on= >HDS terminals for all it's POS operations.  I would bet that = >they run their inventory control, payroll and AR/AP on it as  >well. >   J One of the dirty secrets of the IT world where Object Oriented GUIs, Java K clients and whatnot are deemed a _must_ is that most POS systems still use  G character-based terminals.  Even the companies that sell high-tech use  M decidedly low-tech POS systems (CompUSA, Circuit City, Best Buy, Office Max).   F Why?  Low cost to deploy, low cost to maintain, ease of training (yes,L a simple "green screen" terminal app typically is far easier to train peopleF to use than a GUI app.), faster operation (compare data entry speed onK a character-based app to a GUI app sometime), overall, it's just more cost  J effective and less of a headache.  These advantages aren't limited to POS K systems, but POS systems are so widely deployed that the advantages of the  ) simpler client are more pronounced there.   F This is potentially one of the biggest barriers to Windows everywhere.A Windows just doesn't integrate well with character based apps and A terminals.  Sure, you can put a terminal emulator on Windows, but D why have all the management and cost headaches of Windows at all if G the primary application is the terminal emulator?  It's really too bad  K for Microsoft that they don't have any _real_ vision and insist on pushing  G down our throats "new" technologies like Windows CE and .NET when they  J could have made more inroads in the Enterprise by making an MS-DOS client G architecture that fits in with their present offerings for a low-cost,  * easy-to-maintain character based platform.   >bill  >  >-- K >Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolves E >bill@cs.scranton.edu     |  and a sheep voting on what's for dinner.  >University of Scranton   | B >Scranton, Pennsylvania   |         #include <std.disclaimer.h>      -Jordan Henderson  jordan@greenapple.com    ------------------------------    Date: 15 Jan 2001 11:55:28 -0500* From: young_r@eisner.decus.org (Rob Young)1 Subject: Re: BJ's Warehouse club - using OpenVMS? + Message-ID: <RuWFsvdHw$tY@eisner.decus.org>   \ In article <93v7kq$gfm$1@lisa.gemair.com>, jordan@lisa.gemair.com (Jordan Henderson) writes:. > In article <93v4pi$1v99$3@info.cs.uofs.edu>,/ > Bill Gunshannon <bill@cs.scranton.edu> wrote: = >>One of my favorite local Pizza joints is still using VMS on > >>HDS terminals for all it's POS operations.  I would bet that> >>they run their inventory control, payroll and AR/AP on it as >>well.  >> > L > One of the dirty secrets of the IT world where Object Oriented GUIs, Java M > clients and whatnot are deemed a _must_ is that most POS systems still use  I > character-based terminals.  Even the companies that sell high-tech use  O > decidedly low-tech POS systems (CompUSA, Circuit City, Best Buy, Office Max).  > H > Why?  Low cost to deploy, low cost to maintain, ease of training (yes,N > a simple "green screen" terminal app typically is far easier to train peopleH > to use than a GUI app.), faster operation (compare data entry speed onM > a character-based app to a GUI app sometime), overall, it's just more cost  L > effective and less of a headache.  These advantages aren't limited to POS M > systems, but POS systems are so widely deployed that the advantages of the  + > simpler client are more pronounced there.  > H > This is potentially one of the biggest barriers to Windows everywhere.C > Windows just doesn't integrate well with character based apps and C > terminals.  Sure, you can put a terminal emulator on Windows, but F > why have all the management and cost headaches of Windows at all if I > the primary application is the terminal emulator?  It's really too bad  M > for Microsoft that they don't have any _real_ vision and insist on pushing  I > down our throats "new" technologies like Windows CE and .NET when they  L > could have made more inroads in the Enterprise by making an MS-DOS client I > architecture that fits in with their present offerings for a low-cost,  , > easy-to-maintain character based platform. >     B 	And the best reason (In my opinion) that you mention, is speed.    C 	Where I am at, they have a nice GUI based touch screen in the gift E 	shop.  The girls mostly have these real long fingernails.  To watch  D 	those fingernails in action is a beautiful thing.  They circle the I 	screen, find what they want, and *CLICK*.  Perhaps they would be faster  A 	if they could remember where they need to go next.  But a keypad > 	on a keyboard will always be faster than a keypad on a screen< 	as the angle is all wrong to be effective (i.e. typing on aD 	vertical surface versus typing on a horizontal surface).  So if youF 	are typing on the keypad, who needs a GUI?  They all have the barcodeA 	readers, etc.  So the GUI thing for POS is plain old fluff.  But A 	fast food is even more specialized.  Each key maps to a product. " 	That is even further dumbed down.  6 	The line is always long in this gift shop (sells soda? 	and candy) regardless of who is working the counter.  But they   	have that nice GUI based POS!!!  = 	BTW , slashdot.org ran a good article about the Linux win at @ 	Home Depot.  80K+ POS terminals at Home Depot.  I would be veryC 	worried indeed about Windows ever making inroads in POS.   Someone G 	within Microsoft has to come up with a kiler POS app I suppose.  Maybe C 	if they have a nice spinning roulette wheel that shows us how much C 	of a discount we get on our next purchase... but even at that, you ! 	don't need "fancy" to do that...    				Rob    ------------------------------  # Date: Mon, 15 Jan 2001 12:42:32 GMT  From: Didier.Morandi@gmx.fr  Subject: BMC PAtrol still used? ) Message-ID: <93ur7m$8uj$1@nnrp1.deja.com>   G Browsing through my current Customer's shelves, I've found a BMC Patrol  for ServerWorks demo CD.   Is this product still useful?   A why does the installation procedure ask for a pre-existing PATROL % account that I've never heared about?   5 Where can I find the "Patrol on OpenVMS users guide"?    Thanks,  D.     Sent via Deja.com  http://www.deja.com/   ------------------------------  # Date: Mon, 15 Jan 2001 13:01:06 GMT / From: Mike Price <mike.price@littlewoods.co.uk> # Subject: Re: BMC PAtrol still used? ) Message-ID: <93usaf$9i6$1@nnrp1.deja.com>   E As it is used as a standard on our Unix boxes we have just started to D put it on the VMS machines as well so that we have the same tool all< over. It ceratinly seems to work and seems to be very useful   Mike --B All opinions are my own and do not necessarily reflect those of my employer     Sent via Deja.com  http://www.deja.com/   ------------------------------  % Date: Mon, 15 Jan 2001 11:27:35 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.br # Subject: Re: BMC PAtrol still used? L Message-ID: <OFD68C2C3E.3E927F7A-ON032569D5.004F17E7@ep-bc.petrobras.com.br>  I Try to find something at www.bmc.com. the company I work for will begin a  project J to use Patrol for OpenVMS in a few months.... if the company dont change !  F We have Tivoli and, Candle but these products are not prepared for OVS   Regards    FC l   Two links of PDFs h http://www.bmc.com/rs-bin/RightSite/getcontent/bmcdoc.html?dmw_objectid=09003201801f0aaa&dmw_format=html  h http://www.bmc.com/rs-bin/RightSite/getcontent/bmcdoc.html?dmw_objectid=0900320180223aaf&dmw_format=html        , Didier.Morandi@gmx.fr em 15/01/2001 10:42:32             Info-VAX@Mvb.Saic.Com        Assunto: BMC PAtrol still used?     G Browsing through my current Customer's shelves, I've found a BMC Patrol  for ServerWorks demo CD.   Is this product still useful?2  A why does the installation procedure ask for a pre-existing PATROLa% account that I've never heared about?t  5 Where can I find the "Patrol on OpenVMS users guide"?W   Thanks,u D.     Sent via Deja.coms http://www.deja.com/   ------------------------------  % Date: Mon, 15 Jan 2001 09:03:02 -0600?1 From: "Dave Gudewicz" <david.gudewicz@abbott.com>u Subject: Re: CHARON-VAX 8 Message-ID: <93v3c9$ob8$1@fizban.fizban.pprd.abbott.com>  K At our 1/3/2001 DECUS (ooops Encompass) meeting in Chicagoland, we saw withuI our own eyes a Intel laptop running the Charon VMS emulator.  Neat stuff.   I And we were told it didn't take too much effort to get it up and running.d' In otherwords, even I could do it.  ;-)i   Dave...d  - "Hans Vlems" <hvlems@iae.nl> wrote in message0" news:93sg29$3n7$1@news.IAEhv.nl...$ > Yes, on a RedHat Linux 6.2 system.4 > The configuration file is pretty straight forward.7 > Load a VMS distribution CD in the drive and >>>B DUA0a
 > on a PC!< > Runs well, but the evaluation version has its limitations.G > Mainly that it does not support the (any) network interface of the PCe > and has no X support.dA > The speed is throttled, on a 350 MHz Pentium II the performanceeB > is about 2 VUPS (I wrote a program that makes a crude estimate).C > Main bugs up to now: an image backup of a system disk to an emptye > drive(file) is notI > bootable, and (2) SMG is not well implemented or the VT100 emulation isf wayp > off. >eK > Other than that, it's an amusing tool at home and a good idea if you want  toD > replace VAX hardware. BTW, Unisys does something like that for its A-series@ > mainframes. The low end systems run in software on top of WNT. >n > Hans Vlems >lE > IdrEASY heeft geschreven in bericht <93qs6b$393d$1@as121.tel.hr>...g% > >Was anybody try this VAX emulator?m > >g > >I can't boot it.  > >o > >  >b >    ------------------------------  % Date: Mon, 15 Jan 2001 13:41:31 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.brN Subject: Re: CHARON-VAXoL Message-ID: <OFED05AB12.5B7416E6-ON032569D5.005AD649@ep-bc.petrobras.com.br>  7 Why Compaq dont open the code of the old MicroVMS ! ! ! + Would be better a full port for Intel .....f   Regards    FC        B "Dave Gudewicz" <david.gudewicz@abbott.com> em 15/01/2001 13:03:02             Info-VAX@Mvb.Saic.Comi       Assunto: Re: CHARON-VAXe    K At our 1/3/2001 DECUS (ooops Encompass) meeting in Chicagoland, we saw withtI our own eyes a Intel laptop running the Charon VMS emulator.  Neat stuff.C  I And we were told it didn't take too much effort to get it up and running.-' In otherwords, even I could do it.  ;-)-   Dave...-  - "Hans Vlems" <hvlems@iae.nl> wrote in message " news:93sg29$3n7$1@news.IAEhv.nl...$ > Yes, on a RedHat Linux 6.2 system.4 > The configuration file is pretty straight forward.7 > Load a VMS distribution CD in the drive and >>>B DUA0m
 > on a PC!< > Runs well, but the evaluation version has its limitations.G > Mainly that it does not support the (any) network interface of the PCi > and has no X support.oA > The speed is throttled, on a 350 MHz Pentium II the performance B > is about 2 VUPS (I wrote a program that makes a crude estimate).C > Main bugs up to now: an image backup of a system disk to an emptyh > drive(file) is notI > bootable, and (2) SMG is not well implemented or the VT100 emulation isc wayi > off. >lK > Other than that, it's an amusing tool at home and a good idea if you wanta toD > replace VAX hardware. BTW, Unisys does something like that for its A-series@ > mainframes. The low end systems run in software on top of WNT. >u > Hans Vlems >mE > IdrEASY heeft geschreven in bericht <93qs6b$393d$1@as121.tel.hr>...n% > >Was anybody try this VAX emulator?o > >  > >I can't boot it." > >a > >X >p >e   ------------------------------  % Date: Mon, 15 Jan 2001 09:18:37 +0000 - From: Martin Walker <Martin.Walker@csf.co.uk>R+ Subject: RE: Cluster Alias under Decnet OSI Q Message-ID: <07EC79670CCAD4118F17006097D7C7A70B9B36@london_exch.london.csf.co.uk>t  F Base License Option Numbers for DECnet for OpenVMS VAX are as follows: From the DEcnet SPD...   Alpha End System    QL-MTFA*-AAh  $ Alpha Extended Function  QL-MTGA*-AA  < Alpha - End System to Extended Function Upgrade  QL-MTHA*-AA     -----Original Message-----4 From: John Nixon [mailto:jorlnixon@worldnet.att.net] Sent: 09 January 2001 19:01  To: Info-VAX@Mvb.Saic.Comh+ Subject: Re: Cluster Alias under Decnet OSIk    , I am going to try that first sentence again:  E I cannot seem to find a part number for the DVNETEXT license, which I(? believe we need to ENABLE CLUSTER ALIASES ON our AlphaVMS 7.2-1 L system with DECnet Plus installed .  I tried searching Businesslink (or what. remains of it), but could not come up with it.  J I need to know the price to determine whether it is more cost effective to. change the application, or to buy the license.  : "John Nixon" <jorlnixon@worldnet.att.net> wrote in message> news:d3J66.1095$Cy.76802@bgtnsc06-news.ops.worldnet.att.net...G > I cannot seem to find a part number for the DVNETEXT license, which IeD > believe we need to turn our AlphaVMS 7.2-1 system with DECnet PlusI > installed.  I tried searching Businesslink (or what remains of it), buti > could not come up with it. >aF > The only reference I could find on the web page was that DVNETEXT is# > required to enable cluster alias.a >,I > We are moving an application that runs on VAX systems to Alpha systems,- and-J > the application makes a lot of use of Cluster Aliases for load balancing > purposes.  >jC > Does anyone know the list price of DVNETEXT for all three classesi+ > (workgroup, departmental and enterprise)?p >o >l    A This e-mail including any attachments is confidential and may be aF legally privileged. If you have received it in error please advise the@ sender immediately by return email and then delete it from your  system. B The unauthorised use, distribution, copying or alteration of this F email is strictly forbidden. If you need assistance please contact the& CSF help desk on (+44)(0)207 490 2727.   ------------------------------  % Date: Mon, 15 Jan 2001 11:17:18 +0000 0 From: andrew harrison <andrew.nospam@uk.sun.com>= Subject: Re: Compaq: A simple, affordable clustering solution * Message-ID: <3A62DC3E.5ECA68C5@uk.sun.com>   "David J. Dachtera" wrote: >  > andrew harrison wrote:
 > > [snip]9 > > Don't you think that deeds speek rather more strongly  > > than words.m > >h > > regards. > > Andrew Harrisont > > Enterprise IT ArchitectE > * > Remember, folks: Andrew said it HIMSELF! > ! > ...and you heard it here first!9 >   , I take this reponse as an admission that you lost the argument. I   Regards9 Andrew Harrison. Enterprise IT Architectx   ------------------------------  % Date: Mon, 15 Jan 2001 14:52:14 +0000p0 From: andrew harrison <andrew.nospam@uk.sun.com>= Subject: Re: Compaq: A simple, affordable clustering solution * Message-ID: <3A630E9E.26364BD3@uk.sun.com>   "Main, Kerry" wrote: > =    > Bill,p > =>  J > >>>> It's clearly FUD at least in the sense that a company can deploy a=  ? > 32-processor box and choose *not* to tie it into its existing G > domain-structured NT company directory structure but just allow it to : > communicate as an outsider and manage it separately. <<< > =   J > You mean like a separate NT4 domain. Sure, and some other group can do = the 1 > same thing ie setup their own active directory.E > =2  J > What was W2K biggest advantage? Getting rid of all those separate domai= ns > and trusts in NT4. > =l    @ Kerry you can't have your cake and eat it. In another thread youC question Sun's ability to integrate Solaris with NT/Win2000 stating:A that it is really important and a measure that customers will usei= to decide on which platform they select, then you turn aroundi8 and try to suggest that NT/Win2000 arn't important OS's.  B Since you didn't bother to respond to my posting, probably becauseA yet again you got in a bit too deep can we assume that NT/Win2000eC integration isn't really an issue and that the opinions you voice =   ) in this thread are what you really think.   = You also don't seem to be very up-to-date with Compaqs servern	 strategy.d  ) Let me quote you from your own web pages.-  J http://www.compaq.com/products/servers/proliantml770/description.html#fea= tn  / This is the 32 CPU Unisys box re-sold by CompaqO  9 "Reliability and robustness by combining the enhanced CMPt9 architecture from Compaq plus the extensive service and =l  . support behind our Compaq Data Center Program"  9 And then from the Windows DataCenter Server Program pagesl  4 http://www.compaq.com/solutions/datacenter/what.html  : Incedentally no mention of OpenVMS or Tur64 or Tandem here   But instead we havee  1 WARNING OFFENSIVE COMPAQ MARKETING DRIVEL FOLLOWS   @ "Compaq delivers the highest levels of reliability and stabilityB possible through the Datacenter Certification Program and a highly? structured change management process. In addition to the normalCA comprehensive Compaq ProLiant server testing, the servers will bet? rigorously tested with the Microsoft=AE Windows 2000 Datacentern@ Server operating system software and delivered to the you with aF Microsoft certification that the total solution will work together for8 optimum performance and reliability. Compaq will certifyA applications with hardware or software drivers ("kernel-touching"d? applications) in Compaq's Data Center Certification Lab. CompaqeE will also support third-party applications that receive the Microsofts8 Windows 2000 Datacenter logo and have support escalation agreements with Compaq.e  @ High-availability features within ProLiant servers, StorageWorks: systems, StorageWorks Modular Arrays and Enterprise Backup? Solutions, and the VI-compliant ServerNet II interconnect, plus < Windows 2000 Datacenter solutions such as 4-node clustering,0 deliver 24 x 7 access to mission-critical data."  ? And you are advocating that people should not use Win2000 for =g  / anything mission critical !!!!!!!!!!!!!!!!!!!!!e   Regardsi Andrew Harrisone Enterprise IT ArchitectS   ------------------------------  % Date: Mon, 15 Jan 2001 09:13:01 -0600c7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>l= Subject: Re: Compaq: A simple, affordable clustering solution'- Message-ID: <3A63137D.90F9CC91@earthlink.net>d   andrew harrison wrote: >  > "David J. Dachtera" wrote: > >f > > andrew harrison wrote: > > > [snip]; > > > Don't you think that deeds speek rather more strongly  > > > than words.. > > > 
 > > > regards- > > > Andrew Harrison  > > > Enterprise IT Architectm > >n, > > Remember, folks: Andrew said it HIMSELF! > > # > > ...and you heard it here first!s > >D > . > I take this reponse as an admission that you > lost the argument.   Wrong - as usual!m   --   David J. Dachteraw dba DJE Systemsc http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/a  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.l   ------------------------------  % Date: Mon, 15 Jan 2001 09:49:35 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com>i= Subject: RE: Compaq: A simple, affordable clustering solution(N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284C08@kaoexc3.kao.cpqcorp.net>   Andrew .. Andrew ..a  E Please .. this really is getting boring .. I do not have all day to =  respondrG to all of your fud .. but a few points really do need to be responded =t to -  C >>> And you are advocating that people should not use Win2000 for =' anything mission critical !>>>n  G Please provide a direct quote from me that states this or admit it is =b just( you twisting words to suit your own FUD.  G Sigh, one more time - While Sun's response to all requirements is UNIX,s@ Compaq provides a multi-platform offering and all of them have =	 important 6 strengths and weaknesses in their target platforms.=20  C >>> then you turn around and try to suggest that NT/Win2000 arn't = 	 importante OS's.<<<  G Please provide a direct quote from me that states this or admit it is =o just( you twisting words to suit your own FUD.  I >>> can we assume that NT/Win2000 integration isn't really an issue and =i thatC the opinions you voice in this thread are what you really think.>>>h  F Ask any medium-large Customer if integration with their various OS's = that0 they have is not an important issue for them.=20  H Surely, an Enterprise Architect is aware that integration is important = for  these Customers.=20n  * Rest of dribble about Marketing ignored ..   Regards,  
 Kerry Main Senior Consultantd Compaq Canada Inc. Professional Servicesl Voice: 613-592-4660e Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----7 From: andrew harrison [mailto:andrew.nospam@uk.sun.com]I Sent: January 15, 2001 9:52 AM To: Info-VAX@Mvb.Saic.Come= Subject: Re: Compaq: A simple, affordable clustering solutionl     "Main, Kerry" wrote: >=20 > Bill,e >=20I > >>>> It's clearly FUD at least in the sense that a company can deploy =c ae? > 32-processor box and choose *not* to tie it into its existingtG > domain-structured NT company directory structure but just allow it too: > communicate as an outsider and manage it separately. <<< >=20G > You mean like a separate NT4 domain. Sure, and some other group can =  do the1 > same thing ie setup their own active directory.( >=20E > What was W2K biggest advantage? Getting rid of all those separate =  domainst > and trusts in NT4. >=20  @ Kerry you can't have your cake and eat it. In another thread youC question Sun's ability to integrate Solaris with NT/Win2000 statingiA that it is really important and a measure that customers will use = to decide on which platform they select, then you turn around 8 and try to suggest that NT/Win2000 arn't important OS's.  B Since you didn't bother to respond to my posting, probably becauseA yet again you got in a bit too deep can we assume that NT/Win2000-D integration isn't really an issue and that the opinions you voice=20) in this thread are what you really think.t  = You also don't seem to be very up-to-date with Compaqs serverr	 strategy.u  ) Let me quote you from your own web pages.a  I http://www.compaq.com/products/servers/proliantml770/description.html#fe=h at  / This is the 32 CPU Unisys box re-sold by Compaqg  9 "Reliability and robustness by combining the enhanced CMP : architecture from Compaq plus the extensive service and=20. support behind our Compaq Data Center Program"  9 And then from the Windows DataCenter Server Program pages:  4 http://www.compaq.com/solutions/datacenter/what.html  : Incedentally no mention of OpenVMS or Tur64 or Tandem here   But instead we have   1 WARNING OFFENSIVE COMPAQ MARKETING DRIVEL FOLLOWSh  @ "Compaq delivers the highest levels of reliability and stabilityB possible through the Datacenter Certification Program and a highly? structured change management process. In addition to the normalnA comprehensive Compaq ProLiant server testing, the servers will bet? rigorously tested with the Microsoft=AE Windows 2000 Datacenter,@ Server operating system software and delivered to the you with aF Microsoft certification that the total solution will work together for8 optimum performance and reliability. Compaq will certifyA applications with hardware or software drivers ("kernel-touching" ? applications) in Compaq's Data Center Certification Lab. CompaqeE will also support third-party applications that receive the Microsoft 8 Windows 2000 Datacenter logo and have support escalation agreements with Compaq.c  @ High-availability features within ProLiant servers, StorageWorks: systems, StorageWorks Modular Arrays and Enterprise Backup? Solutions, and the VI-compliant ServerNet II interconnect, plust< Windows 2000 Datacenter solutions such as 4-node clustering,0 deliver 24 x 7 access to mission-critical data."  @ And you are advocating that people should not use Win2000 for=20/ anything mission critical !!!!!!!!!!!!!!!!!!!!!e   Regardss Andrew Harrisonl Enterprise IT Architecte   ------------------------------  % Date: Mon, 15 Jan 2001 17:55:08 +0000e0 From: andrew harrison <andrew.nospam@uk.sun.com>= Subject: Re: Compaq: A simple, affordable clustering solutiono* Message-ID: <3A63397C.7F501053@uk.sun.com>   "David J. Dachtera" wrote: >  > andrew harrison wrote: > >D > > "David J. Dachtera" wrote: > > >m > > > andrew harrison wrote: > > > > [snip]= > > > > Don't you think that deeds speek rather more stronglyU > > > > than words.e > > > >C > > > > regardsw > > > > Andrew Harrisont > > > > Enterprise IT Architecte > > >.. > > > Remember, folks: Andrew said it HIMSELF! > > >u% > > > ...and you heard it here first!n > > >o > >e0 > > I take this reponse as an admission that you > > lost the argument. >  > Wrong - as usual!$ >     Wow, back to the school yard !!!  " How about responding to my points.  7 1.	Have there been recent postings in the past alludingt. 	to Oracle internal sources saying that Oracle( 	is really interested in OpenVMS. yes/no  7 2.	Have these postings also appeared regularly over thea 	last couple of years. yes/noe  5 3.	Have there been recent postings complaining about a* 	Oracle sales people not selling oracle on 	OpenVMS. yes/no  3 4.	Have these postings also appeared regularly over  	the last two years. yes/no   8 5.	Have there been recent postings complaining about the, 	availability of Oracle products on OpenVMS. 	yes/noa  3 6.	Have these postings also appeared regularly overa 	the last two years. yes/nod  > Having answered those six questions perhaps you could explain = why you think anything has changed because another unonymous  < source has said that Oracle is really interested in OpenVMS.  : Incedentally Wrong isn't an answer to these questions any  more than it was last time.t   Regardsl Andrew HarrisonI Enterprise IT Architect    ------------------------------  % Date: Mon, 15 Jan 2001 13:35:49 -05001- From: JF Mezei <jfmezei.spamnot@videotron.ca>c= Subject: Re: Compaq: A simple, affordable clustering solution6, Message-ID: <3A6342FF.63C57C55@videotron.ca>   "Main, Kerry" wrote:I > Sigh, one more time - While Sun's response to all requirements is UNIX, J > Compaq provides a multi-platform offering and all of them have important5 > strengths and weaknesses in their target platforms..  N Because you are well connected to Compaq, *YOU* know that Compaq has availableL different products. But since Compaq does not advertise products such as VMSN or Office Server, the general customer population only sees Windows, Exchange,E Office and Intel solutions. Those a bit connected will see the True64-N offerings  and probably learn that True64 is the first to have had clustering.  6 What use is a product if you're not going to sell it ?  L DII-COE only talks baout making a product supported for 20 years. It doesn'tM do anything about SELLING a product. And that is where Compaq has its probleme	 with VMS.t  L I am affraid that I have to side with Mr Harrisson on this one. (Can I get aJ job at Sun ?) Compaq does a terrible job of leveraging its assets to their fullest potential.   ------------------------------   Date: 15 Jan 2001 15:09:44 GMT1 From: bill@triangle.cs.uofs.edu (Bill Gunshannon) ! Subject: Re: Dead VAXstation 3100i, Message-ID: <93v3ro$1v99$2@info.cs.uofs.edu>  ( In article <93sg7i$3rc$1@news.IAEhv.nl>,%  "Hans Vlems" <hvlems@iae.nl> writes:  |> Where is Freiburg situated?7 |> I live in the Netherlands, might pick up some parts.i   Assuming it hasn't moved :-)C south of of Frankfurt, Mannheim, Karlsruhe and Baden-Baden (in thatt' order) on the Frankfurt-Basel Autobahn.S  ? Great place with a really nice sports complex.  I used to go toW skating competitions there.t   bill   -- IJ Bill Gunshannon          |  de-moc-ra-cy (di mok' ra see) n.  Three wolvesD bill@cs.scranton.edu     |  and a sheep voting on what's for dinner. University of Scranton   |A Scranton, Pennsylvania   |         #include <std.disclaimer.h>   0   ------------------------------  % Date: Mon, 15 Jan 2001 20:33:24 +1300b9 From: "Antony Wardle" <antony.wardle@noospammm.met.co.nz>c" Subject: Re: dec 3000 300 crashing1 Message-ID: <6Rx86.1771$FC1.73637@ozemail.com.au>   
 test mem addrt
 STS-MEM-OK test mem ecc? T_ERR_MEM       - ldl_l bcache miss no victim ad Addr: 00200264x Uncorrected error count = 1. 84 FAIL   	 T mem ecca T-ERR-MEM TNF ECC   
 test mem init. T_STS_MEM ok  
 test mem cells   T-STS-MEM ok  
 test mem bctpg T-ERR-MEM - TNF BCTP   any more help?   antony  # PS THanks for your advice thus far.T   Its been interesting.v       >  > > Test mem > >L8 > > T-STS-MEM - LLSC Test : Addr 00200000 FWD Wr 0000000. > > MCHK ID: 14 Logout Frame address: 00108060> > > exc_addr: 000ED784  bui_stat: 00000050  bui_addr: 001081A0@ > > fill_addr:  00054EF0    hi_synd: 00000000  lo_synd : 0000000B > > mm_csr:  00005931     dc_stat: 00000007          va:  001081E8 > >oA > > ? T-ERR-MEM   - ldl_l bcache miss no victim at Addr: 00200044-* > > T-STS-MEM - Uncorredted Error cont = 1> > > ? T-ERR-MEM _ FAILED status = 20 test Init addr - 00200044 > > ?? 800      MEM 0x0020 >VH > Alas, I don't have service docs so I can't interpret all of this.  The5 MCHK ID in the middle of the memory test looks weird.  >hF > Maybe the problem is not the memory, but rather some external memory circuit. >e$ > From the hardware reference guide: > Memory subtests arei/ >   test mem cell   detecs data stuck at faults-5 >   t mem addr      detects addresses stuck at faults 1 >   t mem llsc      load-locked/store conditionals+ >   t mem bctp      bcache tag parity logicl >   t mem ecc       ECC logic - >   t mem init      inits all memory to zeross$ >   t mem all       all of the above >   t mem ?         help >a > > ?? 800 0020 64mb >v& > This indicates a memory error (duh!)- > The 800 is the FRU code, in the format 8xy.e >    x = bank 0 to 7G >    y = 0 for SIMM 0, 1 for SIMM 1, 8 for both SIMMS, F for can't tell K > (A "bank" of memory is 1 side of a pair of SIMMs.  The other side of both   SIMMs forms another bank. IIRC.) > G > So 800 is supposed to mean SIMM 0 is sick.  Swap the SIMMs, the error  should > change to 801. >.K > The "action" listed for this error is "Be sure all SIMMs are installed int the > > correct postions and are secure.  Rerun the test to verify." >r >: > > PAL code is V5.56 
 > > 64 mb MEMi > , > The firmware that came with VMS 7.1-2 was:< > Console version: V7.0                PALcode version: 5.56 >r > From VMS:n > f$getsyi("console_version")e > f$getsyi("palcode_version")I >L; > But I would NOT try to upgrade firmware with sick memory!  >oB > > So if the memory is faulty, can I take some from out of 1000a?! > > or a VLC, or a 3100, or a PC?c > > 
 > > Antony >-G > Make sure you have the memory configured right.  And it can't hurt tor! remove and re-seat all the SIMMS.u >cL > As far as I know, only the DEC 3000 300 family used this memory.  It isn't/ too hard to find used if you aren't in a hurry.  > ! > There were two DEC memory kits:a* > MS16-BA   16 MB kit (2 SIMMs)     $1,200* > MS16-DA   64 MB kit (2 SIMMs)     $8,000 > 3 > (The prices are from the 1994 DECdirect catalog.)e >T > The catalog description is:t5 > Industry-standard, double-sided 8 MB or 32 MB SIMM.n  > 4 Mbit/16 Mbit DRAM technology > Access time of 70 ns > parity memory technology >e >rH > Clearpoint and Camminton also made compatible kits, and probably other vendors.F > We just bought the last set from clearpoint (used) at work, but Jeff( Sullivan said he thinks he can get more. > L > The systems can hold up to 4 kits.  Each set of 2 SIMMs goes in 2 adjacentJ slots.  You have to load the slots starting with the one farthest from theL drive bay, and don't skip any slots.  If you mix 16 and 64 MB kits, you have to put the big ones in first.r >o > -- > Robert Deininger > rdeininger@mindspring.com    ------------------------------    Date: 15 Jan 2001 12:03:05 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>s" Subject: Re: dec 3000 300 crashingH Message-ID: <y4snmlrsba.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  ; "Antony Wardle" <antony.wardle@noospammm.met.co.nz> writes:e   > test mem eccA > T_ERR_MEM       - ldl_l bcache miss no victim ad Addr: 00200264s > Uncorrected error count = 1n	 > 84 FAIL   I This looks like some circuitry supporting the external, board-level cachebG (hence bcache) is broken. The machine may be able to run without it (at ' certainly severly reduced performance).d   	Jan   ------------------------------  % Date: Mon, 15 Jan 2001 09:47:56 -0000g9 From: "Colin Butcher" <colinDOTbutcherATxdeltaDOTcoDOTuk>a8 Subject: Re: DECNet dyslexia (was: Re: Version dyslexia)A Message-ID: <979552060.12364.0.nnrp-02.9e98f8e9@news.demon.co.uk>E  8 "John Santos" <john.santos@verizon.net> wrote in message8 news:MPG.14cc0c9adad4ba0e9896b7@news.bellatlantic.net...J > In article <979502481.15089.0.nnrp-12.9e98f8e9@news.demon.co.uk>, "Colin6 > Butcher" <colinDOTbutcherATxdeltaDOTcoDOTuk> says...I > > You want to end up with DECnet-Plus V7.2-1-ECO02 for OpenVMS VAX V7.2o (plusi' > > whatever patches you apply to VMS).s > > J > > As I understand it (and as it's working on my VAX VMS V7.2 system) you need+ > > the following in order to achieve this:r > > L > > 1) DECnet-Plus V7.2-1 kit for OpenVMS VAX V7.2 - this was on the layered( > > product CD set (try March 2000 set). >CJ > Just one problem.  I can't do this.  I don't have the CD LP set.  I haveJ > TK50 distribution.  I am under contract.  I never received it.  Are they: > screwing those of us who still have to live with TK50's? >l  G I'd try leaning on the local support centre first, especially as you'rekJ paying for MDDS on TK50 media. They should be able to copy the file from a CD onto TK50 for you.p  J Alternatively someone within Compaq may be able to publish the file on theA patches FTP site to assist anyone else who's been caught by this.a   -- Colinl# (colinDOTbutcherATxdeltaDOTcoDOTuk)m   ------------------------------   Date: 15 Jan 2001 15:43:57 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)3 Subject: Re: Faster gzip, thank you Paul Repacholi!r, Message-ID: <93v5rt$s3b@gap.cco.caltech.edu>   In article <dNgER8WCiOLH@mccdev.slac.stanford.edu>, Fairfield@SLAC.Stanford.EDU (Ken Fairfield; SLAC: 650-926-2924; FAX: 926-3515) writes:. >In article <93le66$sam@gap.cco.caltech.edu>, : >    	mathog@seqaxp.bio.caltech.edu (David Mathog) writes: >[...]M >>                                                    ... I also noticed that L >> if I set the initial file allocation size to be the full size of the fileN >> it slowed things down.  That counterintuitive result must occur because theE >> file system nulls out the disk blocks before releasing them to the  >> application.  ... >-G >        You don't have HIGHWATER enabled  on  the  test disk do you?  i  H It was on on the RAMdisk and off on the U2W disk.  Previous testing has D shown that the HIGHWATER overhead is negligible on the small RAMdisk
 I'm using.  + That's why the result was counterintuitive.    David Mathog mathog@seqaxp.bio.caltech.eduo? Manager, sequence analysis facility, biology division, Caltech a   ------------------------------  # Date: Mon, 15 Jan 2001 16:27:01 GMT,- From: goathunter@goatley.com (Hunter Goatley)y3 Subject: FILESERV@WKU: UnZip V5.42 is now availablem/ Message-ID: <3a6320b1.5714166@swen.process.com>   B The Info-ZIP group has released UnZip 5.42.  I've added VMS objectB files to the normal distribution and updated the WKU archives.  ToA generate executables from the supplied binaries, simply unzip thee9 file, SET DEFAULT [.VMS-BINARIES], and execute @LINK.COM.   & There are two distributions available:      -  UNZIP.ZIP <    -  UNZIP.BCK (a VMS backup saveset for those who'd rather:       not use their previous unzip to unzip the new unzip)  ; You can find UnZip 5.42 for VMS on the following sites now:l  ,   http://www2.wku.edu/www/fileserv/unzip.zip*   ftp://ftp.wku.edu/vms/fileserv/unzip.zip2   ftp://ftp.process.com/wku/vms/fileserv/unzip.zip;   http://www.tmk.com/ftp/ftp-wku-edu/vms/fileserv/unzip.zipu6   ftp://ftp.tmk.com/ftp-wku-edu/vms/fileserv/unzip.zip  5 These mirrors should be updated in the next 24 hours:t  E           ftp.vms.stacken.kth.se, under [.MIRRORS.WKU.VMS.FILESERV].  8           ftp.ctrl-c.liu.se, under [.WKU.VMS.FILESERV].            ftp.riken.go.jp :           ftp.vsm.com.au, under kits and kits/decwindows. :           www.vsm.com.au/ftp/, via the WWW instead of FTP.  9 The supplied VMS binaries have been linked as far back asn$ VAX/VMS V5.4-2 and OpenVMS AXP V1.5.  r Hunter ------9 Hunter Goatley, Process Software, http://www.process.com/r9 goathunter@goatley.com     http://www.goatley.com/hunter/a   ------------------------------  % Date: Mon, 15 Jan 2001 13:30:12 +0000r0 From: andrew harrison <andrew.nospam@uk.sun.com>$ Subject: Re: GS160 hardware question* Message-ID: <3A62FB64.BCCF9086@uk.sun.com>   Fred Kleinsorge wrote: > D > andrew harrison wrote in message <3A5DB184.3A64FBB6@uk.sun.com>... > >"Main, Kerry" wrote:m > >> > >> Wolfgang, > >>K > >> You and the others in this thread might be interested in the following  > WP's:a? > >> <http://www.compaq.com/alphaserver/gs/gs_whitepapers.html>b > >> > >s > >Ohh Kerry really. > >e >  > Random crap removed. >   @ No Fred random truth deleted. Sorry lets just examine the sorry = per CPU memory bandwidth hype contained in the WildFire whitek paper.  ? The White paper claims that one of the benefits of the WildFiretB memory architecture "over other competative systems" is that each @ CPU in a Quad gets 1.6 GB/s of available bandwidth with a total   bandwidth per QUAD of 6.4 GB/s.   ; However sadly for the marketeers some techies went off and s: published the STREAMS numbers for WildFire and they tell a totally different story.  7 Currently the single CPU memory bandwidth is ~1GB/s andn9 this may increase with faster Alpha CPU's. However the 4  9 CPU results show that the "1.6 GB/s per CPU with 32 CPU'so< is bullshit, the limit appears to be ~2.5 GB/s per QUAD not + >6 GB/s this results in ~620 MB/s per CPU. m  ; QBB bandwith is way less than the claims and is not 4 GB/s t? which it should have been if the marketing bs had been correct.k  > Now we know from previous rubbish posted by Compaqs marketing 9 department that maths isn't one of their fortes but 1x4=4a not 2.5.  : In addition if Compaqs marketing claims had been remotely = true, then the all local STREAMS benchmark COMPAQ ran should  ; have got arround 32 GB/s with 32 CPU's they didn't they got 	 ~20 GB/s.e  B This is only one example in a white paper littered with statementsB either disproved by other publications from Compaq or numbers like? the 600ns attributed to Sun's E10K memory latency which appearswA to simply have been made up to make the WildFire memory subsysteme look OK.  N > The truth is that *any* non-uniform memory system will have some penalty forL > access to non-local memory.  NUMA systems have had a wide variety of costsM > and problems, a 3:1 ratio to memory (not already in your cache) is a fairlyoL > good for NUMA systems.  The only good thing to be said for NUMA at all, isL > that it makes it feasible to build systems with large numbers of CPUs.  IfJ > the performance across the switch is faster than the existing system-busF > based design, and the ratio is small enough between local and remoteD > access - then "most" software can simply ignore it (of course someK > software - like locks - do need to deal with it).  If you want to get themL > absolute best performance out of *any* NUMA system - regardless of vendor,N > then you have to be prepared to figure out how to partition your application# > to minimize cross-switch traffic.p >   F The truth is that 3:1 isn't that good, it was good when SGI introduced
 the OriginG 2000 all those years ago. But it isn't that good now. The SuperDome hasn better e@ than 2:1 memory latency with a lower local idle latency than the	 WildFire.uE The Origin 3000 also has a 2:1 memory latency ratio as well also withr better local latency than WildFire.  G However if you do have a memory subsystem with 3:1 ratios between locald and remote rH then you are right you do need to figure out how to partition your data.  G Though why you have refrained from posting this nugget of wisdom before.D is intriguing, Kerry has been arguing for a long time on this group, alongsC with other OpenVMS boosters that any suggestion that data has to be  partitioned H on WildFire to get good performance is just FUD. Where were you ????????  K > Tru64 UNIX has done significant work for these systems (page replication,eK > memory affinity, application partitions, "fair" locks, etc), VMS has donen> > some of the same work for VMS - and more is in the pipeline.   -- r Andrew Harrisona Enterprise IT Architect    ------------------------------    Date: 15 Jan 2001 10:16:30 -0500* From: young_r@eisner.decus.org (Rob Young)$ Subject: Re: GS160 hardware question+ Message-ID: <ijUxZ9TDhcFu@eisner.decus.org>   [ In article <3A62FB64.BCCF9086@uk.sun.com, andrew harrison <andrew.nospam@uk.sun.com writes:   cL > The truth is that 3:1 isn't that good, it was good when SGI introduced theP > Origin 2000 all those years ago. But it isn't that good now. The SuperDome hasJ > better  than 2:1 memory latency with a lower local idle latency than theL > WildFire. The Origin 3000 also has a 2:1 memory latency ratio as well also* > with better local latency than WildFire. >     @ 	Andrew... "3:1 isn't that good".  Depends.  If the numbers for > 	"1" stay the same , then yes 3:1 isn't very good at all.  But9 	from Pete Bannon's presentation from MPR forum on 21364:e 		, 		"Alpha 21364:  A Scalable Single-chip SMP"  , 	we see he is describing these as latencies:   			30 ns CAS latency per pin' 			15 ns processor to processor latency   @ 	now surely there is signalling overhead and whatnot but we also> 	see that the Memory Controller can load lines directly to L1.G 	The point is we will see much better latencies in 21364 and as others -H 	are hard at work integrating memory controllers on-chip (to hide RDRAM H 	latencies).  The on-chip memory controller makes RDRAM very attractive.  M > However if you do have a memory subsystem with 3:1 ratios between local and,L > remote  then you are right you do need to figure out how to partition your > data.X 	   B 	If the numbers look like 120:360 , yes.  But if the numbers startD 	looking like 45:100 then not a big deal right?  Maybe two hops away< 	adds 30 ns and 45:130 is still very attractive.  We'll see.   				Rob    ------------------------------  % Date: Mon, 15 Jan 2001 10:07:16 -0600S+ From: "Main, Kerry" <Kerry.Main@compaq.com>l$ Subject: RE: GS160 hardware questionN Message-ID: <910612C07BCAD1119AF40000F86AF0D805284C0A@kaoexc3.kao.cpqcorp.net>   Andrew,l  @ It looks like after a slow start, you are back in gear again :-)  K >>> Though why you have refrained from posting this nugget of wisdom beforenJ is intriguing, Kerry has been arguing for a long time on this group, alongC with other OpenVMS boosters that any suggestion that data has to bemK partitioned on WildFire to get good performance is just FUD. Where were you  ????????<<<p  B Please provide a direct quote from me that even remotely indicatesI applications do not need to be reviewed for NUMA considerations or simply : admit you are throwing out FUD to suit your own purposes.   I Taking it one step further, ALL applications should be reviewed to see ifnK there are things that can be done optimize their performance when moving to  a new HW platform.  ( Sigh .. nice try, but you can do better.  B Looking forward to you providing me with some past quotes of mine.   :-)f   Regards,  
 Kerry Main Senior Consultantt Compaq Canada Inc. Professional Servicesn Voice: 613-592-4660h Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----7 From: andrew harrison [mailto:andrew.nospam@uk.sun.com]m Sent: January 15, 2001 8:30 AM To: Info-VAX@Mvb.Saic.Comn$ Subject: Re: GS160 hardware question     Fred Kleinsorge wrote: > D > andrew harrison wrote in message <3A5DB184.3A64FBB6@uk.sun.com>... > >"Main, Kerry" wrote:- > >> > >> Wolfgang, > >>K > >> You and the others in this thread might be interested in the following  > WP's:R? > >> <http://www.compaq.com/alphaserver/gs/gs_whitepapers.html>i > >> > >  > >Ohh Kerry really. > >K >  > Random crap removed. >   @ No Fred random truth deleted. Sorry lets just examine the sorry = per CPU memory bandwidth hype contained in the WildFire whitem paper.  ? The White paper claims that one of the benefits of the WildFire B memory architecture "over other competative systems" is that each @ CPU in a Quad gets 1.6 GB/s of available bandwidth with a total   bandwidth per QUAD of 6.4 GB/s.   ; However sadly for the marketeers some techies went off and e: published the STREAMS numbers for WildFire and they tell a totally different story.  7 Currently the single CPU memory bandwidth is ~1GB/s and09 this may increase with faster Alpha CPU's. However the 4 :9 CPU results show that the "1.6 GB/s per CPU with 32 CPU'sv< is bullshit, the limit appears to be ~2.5 GB/s per QUAD not + >6 GB/s this results in ~620 MB/s per CPU. t  ; QBB bandwith is way less than the claims and is not 4 GB/s s? which it should have been if the marketing bs had been correct.e  > Now we know from previous rubbish posted by Compaqs marketing 9 department that maths isn't one of their fortes but 1x4=4  not 2.5.  : In addition if Compaqs marketing claims had been remotely = true, then the all local STREAMS benchmark COMPAQ ran should  ; have got arround 32 GB/s with 32 CPU's they didn't they gote	 ~20 GB/s.   B This is only one example in a white paper littered with statementsB either disproved by other publications from Compaq or numbers like? the 600ns attributed to Sun's E10K memory latency which appearseA to simply have been made up to make the WildFire memory subsystemm look OK.  J > The truth is that *any* non-uniform memory system will have some penalty for L > access to non-local memory.  NUMA systems have had a wide variety of costsF > and problems, a 3:1 ratio to memory (not already in your cache) is a fairlyL > good for NUMA systems.  The only good thing to be said for NUMA at all, isL > that it makes it feasible to build systems with large numbers of CPUs.  IfJ > the performance across the switch is faster than the existing system-busF > based design, and the ratio is small enough between local and remoteD > access - then "most" software can simply ignore it (of course someK > software - like locks - do need to deal with it).  If you want to get theVL > absolute best performance out of *any* NUMA system - regardless of vendor,B > then you have to be prepared to figure out how to partition your applicationy# > to minimize cross-switch traffic.o >   F The truth is that 3:1 isn't that good, it was good when SGI introduced
 the OriginG 2000 all those years ago. But it isn't that good now. The SuperDome hasl better S@ than 2:1 memory latency with a lower local idle latency than the	 WildFire.tE The Origin 3000 also has a 2:1 memory latency ratio as well also withe better local latency than WildFire.  G However if you do have a memory subsystem with 3:1 ratios between local  and remote iH then you are right you do need to figure out how to partition your data.  G Though why you have refrained from posting this nugget of wisdom beforebD is intriguing, Kerry has been arguing for a long time on this group, alongrC with other OpenVMS boosters that any suggestion that data has to beo partitioned H on WildFire to get good performance is just FUD. Where were you ????????  K > Tru64 UNIX has done significant work for these systems (page replication,tK > memory affinity, application partitions, "fair" locks, etc), VMS has dones> > some of the same work for VMS - and more is in the pipeline.   --   Andrew Harrisonx Enterprise IT Architecta   ------------------------------    Date: 15 Jan 2001 11:26:29 -0500/ From: jordan@lisa.gemair.com (Jordan Henderson)-$ Subject: Re: GS160 hardware question* Message-ID: <93v8bl$hf7$1@lisa.gemair.com>  N In article <910612C07BCAD1119AF40000F86AF0D805284C0A@kaoexc3.kao.cpqcorp.net>,* Main, Kerry <Kerry.Main@compaq.com> wrote: >Andrew, >hA >It looks like after a slow start, you are back in gear again :-)s > L >>>> Though why you have refrained from posting this nugget of wisdom beforeK >is intriguing, Kerry has been arguing for a long time on this group, alongrD >with other OpenVMS boosters that any suggestion that data has to beL >partitioned on WildFire to get good performance is just FUD. Where were you >????????<<< >dC >Please provide a direct quote from me that even remotely indicates0J >applications do not need to be reviewed for NUMA considerations or simply; >admit you are throwing out FUD to suit your own purposes. s >-J >Taking it one step further, ALL applications should be reviewed to see ifL >there are things that can be done optimize their performance when moving to >a new HW platform.. > ) >Sigh .. nice try, but you can do better.8 >nC >Looking forward to you providing me with some past quotes of mine.u >s   Don't hold your breath, Kerry.  C This is one of Andrew's weakest points.  He makes all sorts of wildpC accusations about what others have said, without being able to back ! them up with a shred of evidence.y  C I'm still waiting for him to show how it was I who use the term FUDC@ inconsistently, as he claimed multiple times.  This after I was ? able to demonstrate with his own quotes that he was the one whom, had more than one definition for the term.    @ What's amazing is that he still posts to this group at all afterA it's been shown so clearly that he lacks any sort of credibility.i   >:-) > 	 >Regards,r >e >Kerry Mainl >Senior Consultant >Compaq Canada Inc.y >Professional Services >Voice: 613-592-4660 >Fax  :  819-772-7036e >Email: Kerry.Main@Compaq.comy >p >[snip]n   -Jordan Hendersonn jordan@greenapple.comp   ------------------------------    Date: 15 Jan 2001 11:59:31 -0500* From: young_r@eisner.decus.org (Rob Young)$ Subject: Re: GS160 hardware question+ Message-ID: <kT8s6wOcGD3z@eisner.decus.org>   \ In article <93v8bl$hf7$1@lisa.gemair.com>, jordan@lisa.gemair.com (Jordan Henderson) writes:  E > This is one of Andrew's weakest points.  He makes all sorts of wildnE > accusations about what others have said, without being able to backr# > them up with a shred of evidence.9 > E > I'm still waiting for him to show how it was I who use the term FUD B > inconsistently, as he claimed multiple times.  This after I was A > able to demonstrate with his own quotes that he was the one whot. > had more than one definition for the term.   > B > What's amazing is that he still posts to this group at all afterC > it's been shown so clearly that he lacks any sort of credibility.s >   : 	But credibility has little to do with it.  As we have all, 	have determined over the last 4 or 5 years.   				Rob    ------------------------------  % Date: Mon, 15 Jan 2001 18:03:42 +0000i0 From: andrew harrison <andrew.nospam@uk.sun.com>$ Subject: Re: GS160 hardware question* Message-ID: <3A633B7E.E866BDA7@uk.sun.com>   Rob Young wrote: > ] > In article <3A62FB64.BCCF9086@uk.sun.com, andrew harrison <andrew.nospam@uk.sun.com writes:d > N > > The truth is that 3:1 isn't that good, it was good when SGI introduced theR > > Origin 2000 all those years ago. But it isn't that good now. The SuperDome hasL > > better  than 2:1 memory latency with a lower local idle latency than theN > > WildFire. The Origin 3000 also has a 2:1 memory latency ratio as well also, > > with better local latency than WildFire. > >. > H >         Andrew... "3:1 isn't that good".  Depends.  If the numbers forG >         "1" stay the same , then yes 3:1 isn't very good at all.  ButiB >         from Pete Bannon's presentation from MPR forum on 21364: > < >                 "Alpha 21364:  A Scalable Single-chip SMP" > 5 >         we see he is describing these as latencies:o > 3 >                         30 ns CAS latency per pind> >                         15 ns processor to processor latency > I >         now surely there is signalling overhead and whatnot but we alsofG >         see that the Memory Controller can load lines directly to L1.mO >         The point is we will see much better latencies in 21364 and as others P >         are hard at work integrating memory controllers on-chip (to hide RDRAMQ >         latencies).  The on-chip memory controller makes RDRAM very attractive.  >   D This isn't anything to do with WildFire and it also isn't available.  A I could paraphrase your response as 3:1 isn't that good but don'tb? worry Compaq have yet more white papers that describe yet to be @ available systems that will be better than 3:1. We all know how > dangerous a Compaq white paper is in the wrong hands don't we.  O > > However if you do have a memory subsystem with 3:1 ratios between local andrN > > remote  then you are right you do need to figure out how to partition your	 > > data.e >  > K >         If the numbers look like 120:360 , yes.  But if the numbers startiM >         looking like 45:100 then not a big deal right?  Maybe two hops awaymE >         adds 30 ns and 45:130 is still very attractive.  We'll see.g >   < But Rob the numbers arn't currently 120/360 they are 330/9903 as opposed to say the Origin 3000 which is 175/340.s   Regardsr Andrew Harrison  Enterprise IT Architect.   ------------------------------  % Date: Mon, 15 Jan 2001 18:10:09 +0000e0 From: andrew harrison <andrew.nospam@uk.sun.com>$ Subject: Re: GS160 hardware question* Message-ID: <3A633D01.9DBF8C0D@uk.sun.com>   "Main, Kerry" wrote: > 	 > Andrew,o > B > It looks like after a slow start, you are back in gear again :-) > M > >>> Though why you have refrained from posting this nugget of wisdom beforehL > is intriguing, Kerry has been arguing for a long time on this group, alongE > with other OpenVMS boosters that any suggestion that data has to beeM > partitioned on WildFire to get good performance is just FUD. Where were you.
 > ????????<<<  > D > Please provide a direct quote from me that even remotely indicatesK > applications do not need to be reviewed for NUMA considerations or simply0; > admit you are throwing out FUD to suit your own purposes.a >   @ OK Kerry, so you forgot accusing me of FUD when I told you that > Compaq were unlikely to get good TPC-C performance out of the ? WildFire without resorting to the OPS in a box scheme pioneeredu by Sequent.   A Hint OPS in a box is partitioning the database to get around the r issues of having a NUMA system.n  $ Remember what happened next ????????  * > Sigh .. nice try, but you can do better.  4 I hope you arn't in too much pain, all this sighing , could be a symptom of somthing really nasy.    Regardsu Andrew Harrisono Enterprise IT Architect    ------------------------------    Date: 15 Jan 2001 13:48:46 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen) $ Subject: Re: GS160 hardware question+ Message-ID: <TeLMcq93lGS+@eisner.decus.org>l  ] In article <3A633D01.9DBF8C0D@uk.sun.com>, andrew harrison <andrew.nospam@uk.sun.com> writes:e  B > OK Kerry, so you forgot accusing me of FUD when I told you that @ > Compaq were unlikely to get good TPC-C performance out of the A > WildFire without resorting to the OPS in a box scheme pioneeredu > by Sequent.    Oh, Andrew:r  D 	The highest VMS relational database performance is with Oracle Rdb.  - 	Oracle Parallel Server is for Oracle Oracle.a  A Declaring that an irrelevant product is required certainly countsf as FUD in my book.  N ==============================================================================N Great Inventors of our time: Al Gore -> Internet; Sun Microsystems -> ClustersN ==============================================================================   ------------------------------  % Date: Mon, 15 Jan 2001 16:12:20 +0000a9 From: "Miller, Daniel" <Daniel.Miller@nightfreight.co.uk>s Subject: HIghest file versionwA Message-ID: <5E8A0E4920B0D411B1E900508BFCB24029946A@NF-HOUSE-NT1>r   Hi everyone,  D What is the highest file version allowed in VMS?  We have VMS 7.2-1.   TIA,   Daniel Miller    ------------------------------    Date: 15 Jan 2001 17:28:08 +0100G From: Jan Vorbrueggen <jan@mailhost.neuroinformatik.ruhr-uni-bochum.de>g! Subject: Re: HIghest file version H Message-ID: <y4hf30rd9j.fsf@mailhost.neuroinformatik.ruhr-uni-bochum.de>  M It's still at 32767. There's a 16-bit field, negative values are reserved fort relative versions.   	Jan   ------------------------------    Date: 15 Jan 2001 11:39:15 -0500* From: young_r@eisner.decus.org (Rob Young)! Subject: Re: HIghest file version + Message-ID: <ELpUIsHRheFi@eisner.decus.org>t  } In article <5E8A0E4920B0D411B1E900508BFCB24029946A@NF-HOUSE-NT1>, "Miller, Daniel" <Daniel.Miller@nightfreight.co.uk> writes:e > Hi everyone, > F > What is the highest file version allowed in VMS?  We have VMS 7.2-1. >   
 	32767...   4 	Using DFU, you can run a command that sees how high: 	you are getting on file versions.  DFU is on the freeware 	CD... Here is an example:    1 $ dfu search/version=minimum=20000 sys$sysdevice:<  C $7$DUA0:[xxx0.DATA]xxx_yyy.LOG;32759                            0/0 C $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.JUNK;32766                  1/9eC $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.JUNK;32765                  1/9 C $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.JUNK;32767                  1/9iC $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.OLD;32763                   1/9lC $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.OLD;32764                   1/9aC $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.OLD;32765                   1/9nC $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.OLD;32766                   1/9yC $7$DUA0:[SYS1.MAIL$SERVER]NETSERVER.OLD;32767                   1/9n     				Rob    ------------------------------  % Date: Mon, 15 Jan 2001 10:33:54 -0500e* From: Steve Jones <smj@spamfree.crash.com>2 Subject: HW collection (was Re: Dec 3000 300 help)2 Message-ID: <3A631862.7EFCB2A8@spamfree.crash.com>   Paul Repacholi wrote:   D > You have it easy. KL-10E, 11/70, 11/84, ah, some 11/73s, 3 uV-IIs,G > x uV2000s, n 3100s, 6520, 2 M600s, Prioris, noname, 4400, 4200, 3200.eH > I'm not even going to try to remember how many 11/34s. And a couple of > 8s? > Should be a Rainbow, DECmates, and a Vaxmate here somnewhere.d  H I bow before the might - and weight - of your esteemed collection. Which of$ these see use on an irregular basis?  F And by any chance, has the local power utility assigned you a personal account rep? ;^)   --Steve.  E Steve Jones              ...!uunet!crash.com!smj           Arlington,a Mass.n6 CRASH!! Computing     (any spambots parse bang paths?)  E "Chaos will ensue if the variable i is altered..." - SysV Programmers1 GuideS   ------------------------------  % Date: Mon, 15 Jan 2001 14:55:09 +0200 3 From: "Phillip du Plooy" <itbpjdp@puknet.puk.ac.za>D8 Subject: License - Dec C for OpenVMS AXP and broken link- Message-ID: <979563365.354372@news.puk.ac.za>a   Hallo,  K I want to install Dec C on one of my Alpha's, I have the following license: H Part number QL-MU7AE-AA . How do I know on what type of Alpha I can loadG this (1000, 2100 or ES40 all running OpenVMS V7.2-1) on and what is thea? highest version of C that I can run. The SPD links on the page:w8 http://www5.compaq.com/emea/software/uk/COCOVMS.html nl.L http://www.digital.com/SP2538/SP2538SC.TXT  are broken. Where can I download
 the software?a  	 Thank youl Phillip du Plooy DBA/OpenVMS SysAdmin.t  Potchefstroom University for CHE South AfricaD When dealing with backups apply Riley's law: Murphy was an optimist.   ------------------------------  % Date: Mon, 15 Jan 2001 13:41:47 +0000r  From: steven.reece@quintiles.com< Subject: Re: License - Dec C for OpenVMS AXP and broken linkH Message-ID: <OFBC127C82.09C95407-ON802569D5.004A2C2E@qedi.quintiles.com>  K The "E" in the license indicates that it's a Workgroup class license.  This-# should be usable on machines like :mJ DEC 23xx, DEC25xx, DEC33xx, DEC 34xx, DEC36xx, DEC 37xx, AlphaStation 2xx,J AlphaStation 4xx, AlphaStation 5xx, AlphaServer 3xx, 4xx, 8xx, AlphaServer 1000.:  G IIRC, the ES40 dropped a license class when it replaced the AlphaServeriH 4100 in the product range.  Given that the 4100 was a departmental classH server I would expect that the ES40 is a workgroup class system, but get that verified by your VAR. Steve.          G "Phillip du Plooy" <itbpjdp@puknet.puk.ac.za> on 15-01-2001 12:55:09 PM>   To:   Info-VAX@Mvb.Saic.Com ) cc:    (bcc: Steven Reece/QRED/Quintiles)e  9 Subject:  License - Dec C for OpenVMS AXP and broken linke     Hallo,  K I want to install Dec C on one of my Alpha's, I have the following license:iH Part number QL-MU7AE-AA . How do I know on what type of Alpha I can loadG this (1000, 2100 or ES40 all running OpenVMS V7.2-1) on and what is thea? highest version of C that I can run. The SPD links on the page: 8 http://www5.compaq.com/emea/software/uk/COCOVMS.html nl.C http://www.digital.com/SP2538/SP2538SC.TXT  are broken. Where can Ia download
 the software?u  	 Thank youi Phillip du Plooy DBA/OpenVMS SysAdmin.   Potchefstroom University for CHE South AfricaD When dealing with backups apply Riley's law: Murphy was an optimist.   ------------------------------  + Date: Mon, 15 Jan 2001 09:36:30 +0100 (MET)l& From: Rudolf Wingert <win@fom.fgan.de> Subject: Mhz of a Alpha CPU?6 Message-ID: <200101150832.JAA11325@sinet1.fom.fgan.de>   Hello,  K on older Alphas I did found the Mhz of the CPU with the SHOW LICENSE/CHARGEkG command. But on the new one ES40 I can't find the Mhz with a normal DCL.K command. Anybody there, which know an easy DCL command to find out the Mhz?t   TIA and regards Rudolf Wingert   ------------------------------  % Date: Mon, 15 Jan 2001 22:29:50 +1300o9 From: "Antony Wardle" <antony.wardle@noospammm.met.co.nz>   Subject: Re: Mhz of a Alpha CPU?1 Message-ID: <klz86.1863$FC1.76563@ozemail.com.au>r   try show cpu     antony  3 "Rudolf Wingert" <win@fom.fgan.de> wrote in message 0 news:200101150832.JAA11325@sinet1.fom.fgan.de... > Hello, >O> > on older Alphas I did found the Mhz of the CPU with the SHOW LICENSE/CHARGEI > command. But on the new one ES40 I can't find the Mhz with a normal DCL H > command. Anybody there, which know an easy DCL command to find out the Mhz? >a  > TIA and regards Rudolf Wingert >n   ------------------------------  # Date: Mon, 15 Jan 2001 10:51:18 GMT 1 From: CSABA  HARANGOZO   <csabah@zipworld.com.au>   Subject: Re: Mhz of a Alpha CPU?7 Message-ID: <GCA86.408$h4.46076@nostril.pacific.net.au>n  ' Rudolf Wingert <win@fom.fgan.de> wrote:n > Hello,  M > on older Alphas I did found the Mhz of the CPU with the SHOW LICENSE/CHARGE-I > command. But on the new one ES40 I can't find the Mhz with a normal DCLeM > command. Anybody there, which know an easy DCL command to find out the Mhz?h    > TIA and regards Rudolf Wingert   	$ANALYZE/SYSTEM 	SDA>CLUE  CONFIGm 	...  ( 	You might need some privs to do this... 							Csaba  I    ----------------------------------------------------------------------sE    * Csaba I. Harangozo     |    'To err is human', said the hedgehog1E    * csabah@zipworld.com.au |           as he dismounted a wirebrush.hI    ----------------------------------------------------------------------r;    EARTH::AUSTRALIA:[SYDNEY]HARANGOZO.CSABA;1, delete? [N]:m   ------------------------------   Date: 15 Jan 2001 16:15:12 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)1 Subject: Mistakes were made: was RE: Compaq: etc.p, Message-ID: <93v7mg$s3b@gap.cco.caltech.edu>  | In article <910612C07BCAD1119AF40000F86AF0D805284C00@kaoexc3.kao.cpqcorp.net>, "Main, Kerry" <Kerry.Main@compaq.com> writes:/ >Granted, mistakes have been made in the past.    H I've heard this refrain several times now.  It seems to have appeared inF the Compaq lexicon at roughly the same time as "OpenVMS renaissance".    >Thats history.   G Would that it was ONLY history!!!  Compaq's current level of ineptitudenG is well illustrated by the new OpenVMS education program, which arrivedt- DOA, and is as useless to as it is to Compaq.    >I lived it, andL >readers of this list have heard it hundreds of times about the past issues.  F We're STILL living it!  Digital did not keep key software on the AlphaF platform in general, and OpenVMS in particular, and that rendered the F hardware unusable for many tasks.  Compaq has made some noise but has H effectively done very little to remedy this situation.   The Apache portK was a nice gesture (and maybe it has even displaced the OSU server in a feweE places), but it doesn't offset the fact that the sole supported (sic) G browser for VMS is Netscape 3.03, nor does it limit the downside to the H absence of the vast majority of commercial server applications and everyI commercial single user productivity application, from word processors, to  mathematica, to CAD/CAM etc    Regards,   David Mathog mathog@seqaxp.bio.caltech.edue? Manager, sequence analysis facility, biology division, Caltech e   ------------------------------  # Date: Mon, 15 Jan 2001 09:25:48 GMTt From: pdafniotis@hotmail.comB Subject: Re: OpenGL on a Compaq XP1000 w/ Elsa Gloria Synergy card( Message-ID: <93ufmo$n9$1@nnrp1.deja.com>  , In article <93nfjf$3ur9$1@lead.zk3.dec.com>,8   "Fred Kleinsorge" <kleinsorge@star.zko.dec.com> wrote:D > You need to get a P300 or P350 card for OpenGL support.  Late this yearH > there will be another set of cards with OpenGL support.  The ELSA does not.F > I could tell you how to get Software GL to work, but it is painfully slow.e >eB > If you have to use the ELSA card and want to use GL, I'd get the MesaGL > freeware (and GLUT). >h > _Fredo   Fred and others-  E I have searched the official Compaq docs re P300, P350 and nowhere doiH they mention explicitly OpenVMS. Are these cards guaranteed 100% to work$ with OpenVMS on OpenGL applications?  E Also, could someone give me an idea of the cost we are talking about?a  D I do not need to mention how bad this is for Compaq to advertise theH XP1000 as THE MACHINE for 3-D graphics and not say explicitly the issuesH with ELSA and perhaps with other cards. Still I recognize that this is aG decision taken by someone several levels above Fred and the other folksuG from Compaq who write in this forum; hence, I do REALLY appreciate yours2 help and input on this and I am thankful for this.  
 Kind regards,i   Petros ---e Dr. Petros Dafniotis Senior Research Engineer     Sent via Deja.com  http://www.deja.com/   ------------------------------  % Date: Mon, 15 Jan 2001 15:05:27 +0000m0 From: andrew harrison <andrew.nospam@uk.sun.com>E Subject: Re: OpenVMS and DII COE (was Re: IBM + Linux + US$ 1 bilion)t* Message-ID: <3A6311B7.E6CB36DC@uk.sun.com>   Fred Kleinsorge wrote: > G > Everyone take a deep breath.  If you don't know what COE is, then youiM > probably don't need to know.  If you do, then you probably already know the  > important facts. > M > - There are two COE platform types, one is POSIX-like systems, the other isi
 > Windows. > M > - Only a single vendor has every passed DII/COE certification: Compaq Tru64	 > UNIX.   B Sorry you forgot SGI IRIX which has also passed the COE compliance) testing. Got its stamp in July last year..     > C > - All other implementations pass because they are the "reference"7 > implementations. >   8 HP-UX 10 and Solaris fall into this category so the only1 major UNIX OS missing in action currently is AIX.    Regardsr Andrew Harrisons Enterprise IT Architectd   ------------------------------  % Date: Mon, 15 Jan 2001 11:45:14 -0300 ) From: fabio_compaq@ep-bc.petrobras.com.brr/ Subject: OpenVMS Volume Management (On the fly) L Message-ID: <OF1AEB6326.76A57F18-ON032569D5.0050ABD7@ep-bc.petrobras.com.br>  < Do the RAID 5 sofware or the Volume Shadow  for OpenVMS haveA   online volume management - file system shrinking or growing ???f. As the VERITAS Volume Manager do for Solaris ?  G Or I must dismount all the disks and remount the volume with the add-one disk for the volume grows?   Regards    FC   ------------------------------  % Date: Mon, 15 Jan 2001 09:30:03 -0600e7 From: "David J. Dachtera" <djesys.nospam@earthlink.net> 3 Subject: Re: OpenVMS Volume Management (On the fly) , Message-ID: <3A63177B.6D328DB@earthlink.net>  * fabio_compaq@ep-bc.petrobras.com.br wrote: > > > Do the RAID 5 sofware or the Volume Shadow  for OpenVMS haveC >   online volume management - file system shrinking or growing ??? 0 > As the VERITAS Volume Manager do for Solaris ?   Um, well, none of the above.  tI > Or I must dismount all the disks and remount the volume with the add-onT > disk for the volume grows?  = You'll want to check the doc. set and read up on volume-sets.h  H The only other choice is to physically change the characteristics of theE storage-set presented to VMS by the (HSJ, HSZ, HSG, etc.). Of course,r. that means, BACKUP before and restore after...   --   David J. Dachtera  dba DJE Systems  http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/1  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.h   ------------------------------  % Date: Mon, 15 Jan 2001 13:31:43 -0300r) From: fabio_compaq@ep-bc.petrobras.com.brb3 Subject: Re: OpenVMS Volume Management (On the fly)sL Message-ID: <OFEEE554CD.B44BE638-ON032569D5.005A7B70@ep-bc.petrobras.com.br>  B This is bad, because Sun uses these arguments when they sell their "High Availbility Systems".   A Volume Manager can do file system grow/shrinking on the fly .....h  H I think the storage people from Compaq must re-think the features of theD  Volune Shadow and RAID 5 software. Mount another disk in the volume on the fly .....     Regards    FC        H "David J. Dachtera" <djesys.nospam@earthlink.net> em 15/01/2001 13:30:03             Info-VAX@Mvb.Saic.Com       3 Assunto: Re: OpenVMS Volume Management (On the fly)t    * fabio_compaq@ep-bc.petrobras.com.br wrote: >t> > Do the RAID 5 sofware or the Volume Shadow  for OpenVMS haveC >   online volume management - file system shrinking or growing ??? 0 > As the VERITAS Volume Manager do for Solaris ?   Um, well, none of the above.  I > Or I must dismount all the disks and remount the volume with the add-onu > disk for the volume grows?  = You'll want to check the doc. set and read up on volume-sets.n  H The only other choice is to physically change the characteristics of theE storage-set presented to VMS by the (HSJ, HSZ, HSG, etc.). Of course,o. that means, BACKUP before and restore after...   -- David J. DachteraM dba DJE Systemse http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/l  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.:   ------------------------------  # Date: Mon, 15 Jan 2001 10:18:21 GMT 1 From: CSABA  HARANGOZO   <csabah@zipworld.com.au>a3 Subject: Re: SYSGEN details missing from Audit logsr7 Message-ID: <N7A86.407$h4.46012@nostril.pacific.net.au>   ( Greyskull <Greyskull@infosec.com> wrote:' > Can anyone advise on this one please.t  F > In the audit logs, a sysgen parameter change used to report both theB > old and the new value and the parameter that changed. Now, under > 7.2-1, they've gone.  D > Is there something that isn't set correctly or are VMS engineeringE > taking lessons from Microsoft, the gurus of  useless security logs?=  @ > Apologies if this one has been done to death in the NG alreadyC > while I wasn't watching. Advice on this one would be appreciated.l  A 	I think you have to turn on the the auditing for SYSGEN security=% 	events. It might not be the default.-# 	Please see the help on $SET AUDIT.1  ? 	Also please read the manual, OpenVMS Guide to System Security,c> 	chapter 9, "Security Auditing". You might have a printed form: 	of this manual, or you can find it on a documentation CD.= 	If all else fails, you can read the manual on the Internet :   	http://www.openvms.compaq.com .   	Hope this helps...			Csaba-  I    ----------------------------------------------------------------------rE    * Csaba I. Harangozo     |    'To err is human', said the hedgehogeE    * csabah@zipworld.com.au |           as he dismounted a wirebrush. I    ---------------------------------------------------------------------- ;    EARTH::AUSTRALIA:[SYDNEY]HARANGOZO.CSABA;1, delete? [N]:-   ------------------------------  # Date: Mon, 15 Jan 2001 08:51:43 GMT(" From: pa@it.singer-friedlander.com  Subject: Re: UCX Routing Problem) Message-ID: <93udmt$vdc$1@nnrp1.deja.com>E  F Here is the information that you have requested re the Routing issue :  # $PA(LDN)-> ucx show conf inter/fulli Interface: ZE0C IP_Addr: 129.100.1.11 NETWRK: 255.255.0.0   BRDCST: 129.100.255.255.3 C_Addr:             C_NETWRK:             C_BRDCST:n   Flags: Receive buffer: 0e   Interface: LO03 IP_Addr: 127.0.0.1    NETWRK: 255.0.0.0     BRDCST:x    # $PA(LDN)-> ucx show route/perm/full2  /                              PERMANENT databaseL  < Type           Destination                           Gateway  9 PN  * 0.0.0.0                             * 129.100.203.1.9 PH  * 10.200.10.2                         * 129.100.203.1D     $PA(LDN)-> ucx show route/full  -                              DYNAMIC databaseS  < Type           Destination                           Gateway  F AH  * 127.0.0.1        LOCALHOST          * 127.0.0.1        LOCALHOST9 AH  * 10.200.10.2                         * 129.100.203.1:9 AN  * 0.0.0.0                             * 129.100.203.1y9 AN  * 10.0.0.0                            * 129.100.203.2 @ AN  * 129.100.0.0                         * 129.100.1.11     LDN     Cheers,    Piyush  . In article <t5ufe5ff2lhfa@news.supernews.com>,(   "John Vottero" <John@mvpsi.com> wrote: >02 > "John Vottero" <John@mvpsi.com> wrote in message+ > news:t5ubhvtiuuk01b@news.supernews.com...  > [snip]C > > The default route looks broken anyway.  I don't machine with an3
 address of5 > > 10.x.x.x can't have a default route of 129.x.x.x.  > [snip] > ? > I can't even understand my own post.  What I meant to say wasb	 something, > like:, >d@ > I don't think a machine with an address of 10.x.x.x can have a
 default routee > of 129.x.x.x.W >h >f     Sent via Deja.comr http://www.deja.com/   ------------------------------  % Date: Mon, 15 Jan 2001 12:05:30 -0500 % From: "John Vottero" <John@mvpsi.com>   Subject: Re: UCX Routing Problem/ Message-ID: <t66bg59piktvc8@news.supernews.com>   L Look for the cause of the problem on the 129.100.203.3 machine.  It seems to5 be claiming that it's a path to the 10.0.0.0 network.-  / <pa@it.singer-friedlander.com> wrote in message0# news:93udmt$vdc$1@nnrp1.deja.com...dH > Here is the information that you have requested re the Routing issue : >k% > $PA(LDN)-> ucx show conf inter/fullk > Interface: ZE0E > IP_Addr: 129.100.1.11 NETWRK: 255.255.0.0   BRDCST: 129.100.255.255l5 > C_Addr:             C_NETWRK:             C_BRDCST:a >y > Flags: > Receive buffer: 00 >  > Interface: LO05 > IP_Addr: 127.0.0.1    NETWRK: 255.0.0.0     BRDCST:s >  >C% > $PA(LDN)-> ucx show route/perm/fulla >w1 >                              PERMANENT databaset >n> > Type           Destination                           Gateway >a; > PN  * 0.0.0.0                             * 129.100.203.1 ; > PH  * 10.200.10.2                         * 129.100.203.1t >f >a  > $PA(LDN)-> ucx show route/full >e/ >                              DYNAMIC database- >-> > Type           Destination                           Gateway >nH > AH  * 127.0.0.1        LOCALHOST          * 127.0.0.1        LOCALHOST; > AH  * 10.200.10.2                         * 129.100.203.1E; > AN  * 0.0.0.0                             * 129.100.203.1 ; > AN  * 10.0.0.0                            * 129.100.203.2aB > AN  * 129.100.0.0                         * 129.100.1.11     LDN >, >p	 > Cheers,x >  > Piyush >40 > In article <t5ufe5ff2lhfa@news.supernews.com>,* >   "John Vottero" <John@mvpsi.com> wrote: > >c4 > > "John Vottero" <John@mvpsi.com> wrote in message- > > news:t5ubhvtiuuk01b@news.supernews.com...e
 > > [snip]E > > > The default route looks broken anyway.  I don't machine with anb > address of7 > > > 10.x.x.x can't have a default route of 129.x.x.x."
 > > [snip] > >pA > > I can't even understand my own post.  What I meant to say was  > something:	 > > like:c > >DB > > I don't think a machine with an address of 10.x.x.x can have a > default route, > > of 129.x.x.x.l > >  > >o >o >y > Sent via Deja.comf > http://www.deja.com/   ------------------------------  % Date: Mon, 15 Jan 2001 12:34:50 +0000() From: Christof Brass <brass@infopuls.com>- Subject: Re: Variables in DCL , Message-ID: <3A62EE6A.104BDCBE@infopuls.com>   Tom Gee wrote: > J > Got a bit tricky problem with DCL, can't find out how to solve it... I'mB > trying to write a DCL command procedure where another DCL COM isB > written, and this COM finally creates a third one. I have to useG > variables that can be resolved in a different environment and after a'I > certain time in the second COM. It seems hard to write the first COM sooJ > that the variables come in a right way to the second and third COMs. The# > problem occurs e.g. in this line:g > F > $ write OUT_1 "$ write OUT_2 ""$ if f$loc(""''STRING_2'"", STRING_1)  > .ne. f$len(STRING_1) """ ! <-- > E > How should I write it to get the STRING_2 variable correctly in the " > second COM, something like this: > I > $ write OUT_2 "$ if f$loc(''STRING_2', STRING_1) .ne. f$len(STRING_1) "  > G > This example doesn't necessarily do anything reasonable, it's just to  > make clear the problem:  >  > N6>ty composvms.comd > $! set noverify. > $ set verify > $ STRING_1 = "123ABCDEF" > $ open/write OUT_1 TEST_1.COMm > $o& > $ write OUT_1 "$ STRING_2 = ""ABC""" > $ / > $ write OUT_1 "$ open/write OUT_2 TEST_2.COM"sA > $ write OUT_1 "$ write OUT_2 ""$ write sys$output """"STRING_1:d > ''STRING_1'"""" """hA > $ write OUT_1 "$ write OUT_2 ""$ write sys$output """"STRING_2:h > ''STRING_2'"""" """fC > $ write OUT_1 "$ write OUT_2 ""$ STRING_1 = """"456ABCHIJ"""" """a > $yF > $ write OUT_1 "$ write OUT_2 ""$ if f$loc(""''STRING_2'"", STRING_1)  > .ne. f$len(STRING_1) """ ! <--+ > $ write OUT_1 "$ write OUT_2 ""$ then """ I > $ write OUT_1 "$ write OUT_2 ""$     write sys$output """"Found"""" """l, > $ write OUT_1 "$ write OUT_2 ""$ endif """ > $a > $ write OUT_1 "$ close OUT_2"< > $ close OUT_1w > $ type TEST_1.COM  > $ @TEST_1a > $ type TEST_2.COM  > $ exit >  > N6>@composvmsE > $! set noverify. > $ set verify > $ STRING_1 = "123ABCDEF" > $ open/write OUT_1 TEST_1.COMA > $y& > $ write OUT_1 "$ STRING_2 = ""ABC""" > $d/ > $ write OUT_1 "$ open/write OUT_2 TEST_2.COM"tA > $ write OUT_1 "$ write OUT_2 ""$ write sys$output """"STRING_1:o > 123ABCDEF"""" """.J > $ write OUT_1 "$ write OUT_2 ""$ write sys$output """"STRING_2: """" """ > C > $ write OUT_1 "$ write OUT_2 ""$ STRING_1 = """"456ABCHIJ"""" """d > $h@ > $ write OUT_1 "$ write OUT_2 ""$ if f$loc("""", STRING_1) .ne. > f$len(STRING_1) """ ! <---+ > $ write OUT_1 "$ write OUT_2 ""$ then """-I > $ write OUT_1 "$ write OUT_2 ""$     write sys$output """"Found"""" """m, > $ write OUT_1 "$ write OUT_2 ""$ endif """ > $2 > $ write OUT_1 "$ close OUT_2"q > $ close OUT_1d > $ type TEST_1.COMj > $ STRING_2 = "ABC" > $ open/write OUT_2 TEST_2.COM-= > $ write OUT_2 "$ write sys$output ""STRING_1: 123ABCDEF"" "A4 > $ write OUT_2 "$ write sys$output ""STRING_2: "" "- > $ write OUT_2 "$ STRING_1 = ""456ABCHIJ"" " @ > $ write OUT_2 "$ if f$loc("", STRING_1) .ne. f$len(STRING_1) " > $ write OUT_2 "$ then ".3 > $ write OUT_2 "$     write sys$output ""Found"" "  > $ write OUT_2 "$ endif " > $ close OUT_2" > $ @TEST_1h > $ STRING_2 = "ABC" > $ open/write OUT_2 TEST_2.COM = > $ write OUT_2 "$ write sys$output ""STRING_1: 123ABCDEF"" "e4 > $ write OUT_2 "$ write sys$output ""STRING_2: "" "- > $ write OUT_2 "$ STRING_1 = ""456ABCHIJ"" ",@ > $ write OUT_2 "$ if f$loc("", STRING_1) .ne. f$len(STRING_1) " > $ write OUT_2 "$ then "T3 > $ write OUT_2 "$     write sys$output ""Found"" "t > $ write OUT_2 "$ endif " > $ close OUT_2  > $ type TEST_2.COMr* > $ write sys$output "STRING_1: 123ABCDEF"! > $ write sys$output "STRING_2: "  > $ STRING_1 = "456ABCHIJ". > $ if f$loc(", STRING_1) .ne. f$len(STRING_1) > $ then  > $     write sys$output "Found"	 > $ endif  > $ exit >  > Grateful for any help, >  > Tomf  ; Another simple way to get things simpler is to use templates@ files for each stage. The template DCL files contains a skeleton, with tags you replace by the current values.   ------------------------------  % Date: Mon, 15 Jan 2001 12:35:20 -0500d- From: "Peter Weaver" <peter.weaver@stelco.ca>  Subject: Re: Variables in DCLh4 Message-ID: <JyG86.120294$Z2.1428552@nnrp1.uunet.ca>  / "Tom Gee" <gemtoo@hotmail.com> wrote in message-% news:3A622496.67B2F0D7@hotmail.com...a > ...0F > $ write OUT_1 "$ write OUT_2 ""$ if f$loc(""''STRING_2'"", STRING_1)  > .ne. f$len(STRING_1) """ ! <-- > ...e  D What I always do is start with the procedure you want to end up withD and replace every " with " + quote + " and replace every '' with " +A tick + tick + " then replace every remaining ' with " + tick + ".   E Then insert $ WRITE OUTFILE " at the start of every line and then adde. these two lines at the start of the procedure;       $ quote[0,7] = 34i     $ tick[0,7] = 39  7 figure out the logic for the variables you want to use.,  / Since you are going three levels deep I may use        $ quote2[0,7] = 34     $ tick2[0,7] = 39   D in there so I can keep track of the second level quotes and ticks so you end up with something like;a   $ quote[0,7]=34  $ tick[0,7]=39 $ open/write outfile x.x" $ write outfile "$ quote2[0,7]=34"  $ write outfile "$ tick[0,7]=39"E $ write outfile "$ write sys$output " + quote + "$ write sys$output " 	 + quote +HE " + quote2 + " + quote + "STRING_1: 123ABCDEF" + quote + " + quote2 + 	 " + quote  + "oE $ write outfile "$ write sys$output " + quote + "$ write sys$output " 	 + quote +eF " + quote2 + " + quote + "STRING_2: " + quote + " + quote2 + " + quote + "eA $ write outfile "$ write sys$output " + quote + "$ STRING_1 = " +n quote + " + quC ote2 + " + quote + "456ABCHIJ" + quote + " + quote2 + " + quote + "cE $ write outfile "$ write sys$output " + quote + "$ if f$loc(" + quote-
 + " + quot; e2 + quote2 + " + quote + ", STRING_1) .ne. f$len(STRING_1)w7 $ write outfile "$ write sys$output " + quote + "$ thenS< $ write outfile "$ write sys$output " + quote + "$     write sys$output " + quotuE e + " + quote2 + " + quote + "Found" + quote + " + quote2 + " + quote. + "y8 $ write outfile "$ write sys$output " + quote + "$ endif $ close outfileu     --   RULES OF THE AIR   -----------------t;  #11. The probability of survival is inversely proportionalP<       to the angle of arrival. Large angle of arrival, small-       probability of survival and vice versa.t   ------------------------------   Date: 15 Jan 2001 12:09:11 GMT' From: david20@alpha2.mdx.ac.uk (D.Webb)G< Subject: Re: Virus scanning on VMS (was Re: OpenVMS on CNET)0 Message-ID: <93up97$imk$1@aquila.news.mdx.ac.uk>  _ In article <87bstccatq.fsf@k9.prep.synonet.com>, Paul Repacholi <prep@prep.synonet.com> writes:i4 >nothome@spammers.are.scum (Malcolm Dunnett) writes: >rS >>     It's been my experience that Sophos is at least as fast as McAfee at gettingaQ >> new virus identities released. I have an automated procedure set up here whichS@ >> automatically downloads new virus identities every few hours. >0  >I should not be supprised, but. >bH >How often, in you opinion, do you need to update your scan templates toH >have a resonable level of security on the gateshit? Any idea on capital >and operating costs?r >l  0 If you goto http://www.sophos.com/downloads/ide/( you can see the latest list of IDE filesE (IDEntity files used to identify new viruses). This page includes the + date on which the IDE files were released.    M I haven't actually automated this myself yet. You can register to be notifiedaN by email when there is a new identity file available and then I or a colleagueH just download it. I ought to be able to automate this using deliver but  haven't got round to it.  L As to costs we have an unlimited site license (which includes use at home by< members of the University) which costs about 3000 per year.    
 David Webb VMS and Unix team leader CCSS Middlesex University   >-- = >Paul Repacholi                               1 Crescent Rd.,c8 >+61 (08) 9257-1001                           Kalamunda.A >                                             West Australia 6076t/ >Raw, Cooked or Well-done, it's all half baked.    ------------------------------  % Date: Mon, 15 Jan 2001 16:27:38 +0800 + From: Steven Xie <r33300@email.sps.mot.com>w Subject: vms tooln1 Message-ID: <3A62B47A.48FCBBC5@email.sps.mot.com>o  , This is a multi-part message in MIME format.& --------------BF0F14BCD6728016F2702C40* Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bitg  F Hello there, is there a tool for OpenVMS, it can split a big file into small size?r   Thx. Steven  & --------------BF0F14BCD6728016F2702C40- Content-Type: text/x-vcard; charset=us-ascii;1  name="r33300.vcf" Content-Transfer-Encoding: 7bitm( Content-Description: Card for Steven Xie  Content-Disposition: attachment;  filename="r33300.vcf"   begin:vcard  n:Xie;Steven tel;pager:191/192-296-2212 tel;work:2370-0131 x-mozilla-html:FALSEL org:<img src=http://nvmtc.sps.mot.com/images/animated_motorola.gif>;MOS17 IT version:2.1 ' email;internet:r33300@email.sps.mot.comp title:OpenVMS Administratort adr;quoted-printable:;;MOTOROLA CHINA ELECTRONICS LTD=0D=0AXI QING SITE, MOS 17=0D=0ANO. 15, XING HUA ROAD,=0D=0AXIQING ECONOMIC DEVELOPMENT AREA;Tianjin;;300381;Chinam x-mozilla-cpt:;-142881
 fn:Steven Xie 	 end:vcard   ( --------------BF0F14BCD6728016F2702C40--   ------------------------------    Date: 15 Jan 2001 08:11:02 -05009 From: Kilgallen@eisner.decus.org.nospam (Larry Kilgallen)  Subject: Re: vms toole+ Message-ID: <Ync8QlgzDFRB@eisner.decus.org>s  _ In article <3A62B47A.48FCBBC5@email.sps.mot.com>, Steven Xie <r33300@email.sps.mot.com> writes:   H > Hello there, is there a tool for OpenVMS, it can split a big file into
 > small size?k  5 $ BACKUP/VERIFY BIGFILE.TXT DVA0:SAVESET.BCK/SAVE :-)e   ------------------------------  % Date: Mon, 15 Jan 2001 09:24:37 -0600a7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>V Subject: Re: vms toolG- Message-ID: <3A631635.D8DE0B17@earthlink.net>    Steven Xie wrote:  > H > Hello there, is there a tool for OpenVMS, it can split a big file into
 > small size?p >  > Thx. > Steven  B I have a couple of DCL proc.'s that can do this, but only for .ZIPG and/or .TGZ files or other Fixed-512 files. I can post them, if they'll  help...    -- c David J. Dachteraf dba DJE Systemsp http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/-  F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.m   ------------------------------    Date: 15 Jan 2001 10:42:39 -0500* From: young_r@eisner.decus.org (Rob Young) Subject: Re: vms toolh+ Message-ID: <ev1c9u9REo+m@eisner.decus.org>o  _ In article <3A62B47A.48FCBBC5@email.sps.mot.com>, Steven Xie <r33300@email.sps.mot.com> writes:e > H > Hello there, is there a tool for OpenVMS, it can split a big file into
 > small size?f >   I 	There is a tool called SPLOIN.  I don't see it on the freeware CD.  Herev 	is a look at it:l  = LINK.COM;1                 1/9        29-AUG-1994 15:50:25.47n= README.FIRST;1             1/9         3-SEP-1994 00:02:20.00 = SPLOIN.;1                 38/45        3-SEP-1994 00:02:20.00a= SPLOIN.ALPHA_OBJ;1        93/99        6-SEP-1994 16:33:43.94 = SPLOIN.C;1                47/54        3-SEP-1994 00:02:20.00o= SPLOIN.DOC;1              38/45        3-SEP-1994 00:02:20.00a= SPLOIN.OBJ;1              25/27        6-SEP-1994 16:34:12.87q SPLOIN.PRODUCT-INFO;1l=                            3/9         3-SEP-1994 00:02:20.00-= VMS.ALPHA_OBJ;1            9/9         6-SEP-1994 16:33:49.95'= VMS.C;1                    5/9         3-SEP-1994 00:02:20.00a= VMS.OBJ;1                  3/9         6-SEP-1994 16:34:12.87-= VMSMAKE.COM;1              1/9        29-AUG-1994 15:42:05.14R  " 	Here is some of the product info:   .descriptionI Sploin is a highly configurable file splitter and joiner. Can extract any L number of bytes from anywhere in a file, separate a file in two (with headerN skiping), split a file in smaller ones (specifying the number of chunks or theK size of each) with optional auto-truncation to MS-DOS 8.3 chars limitation,cK and joining with various ways of specifying the files. The major feature is J it's ability to split a big file into smaller ones, and then automaticallyL joining them together later. Can be compiled on practically all UNIX and VMSC systems (thus ideal for bringing home huge archives freshly FTPed).  .version 1.79 .dateL
 1994.09.02 .authors Yves Perrenoud
 .requirements  A computer with a C compiler
 .distribution  Freeware .address Yves Perrenoud Traversiere 29 CH-2013 Colombier  SwitzerlandD .email pyves@alphanet.cho pyves@nuga.alphanet.ch pyves@neli00.eicn.etna.cht Yves.Perrenoud@eicn.etna.chi pyves@litamiga.epfl.ch  ; 	I do recall using it and it does work.  Didn't find a real C 	need for it so I gave up on doing something with it.  If you can't @ 	find it on the net, and can't contact the author... I'll try to@ 	send it to you as an attachment (an Info-ZIP file of the source 	and .com files)..   				Rob    ------------------------------   Date: 15 Jan 2001 16:52:33 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog) Subject: Re: vms toolc, Message-ID: <93v9sh$s3b@gap.cco.caltech.edu>  _ In article <3A62B47A.48FCBBC5@email.sps.mot.com>, Steven Xie <r33300@email.sps.mot.com> writes:  >kG >Hello there, is there a tool for OpenVMS, it can split a big file intot >small size?  ? Depends what you mean by "split".  But I think probably EXTRACTo? will do what you want - it will yank out bits by record, block, - or column.  Pick it up from the WKU site via:.  0   http://www2.wku.edu/www/fileserv/fileserv.html     David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech 3   ------------------------------    Date: 15 Jan 2001 13:04:38 -0500* From: young_r@eisner.decus.org (Rob Young) Subject: Re: vms toolo+ Message-ID: <pEAbnEIhdA2G@eisner.decus.org>0  a In article <93v9sh$s3b@gap.cco.caltech.edu>, mathog@seqaxp.bio.caltech.edu (David Mathog) writes:-a > In article <3A62B47A.48FCBBC5@email.sps.mot.com>, Steven Xie <r33300@email.sps.mot.com> writes:  >>H >>Hello there, is there a tool for OpenVMS, it can split a big file into
 >>small size?  > A > Depends what you mean by "split".  But I think probably EXTRACT2A > will do what you want - it will yank out bits by record, block,T/ > or column.  Pick it up from the WKU site via:e > 2 >   http://www2.wku.edu/www/fileserv/fileserv.html >  >   = 	Sploin is found there too.  Which begs the question.... what 9 	are you trying to do?  Split files into smaller files or  	extract portions of files?    				RobG   ------------------------------  # Date: Mon, 15 Jan 2001 16:23:30 GMT  From: waybright@my-deja.com 2 Subject: What is the installed base of Tru64 Unix?) Message-ID: <93v85n$k51$1@nnrp1.deja.com>   @ It's well advertised that the installed base for VMS is 450,000+H systems.  Does anyone have figures on the installed base for Tru64 UNIX?     Sent via Deja.com  http://www.deja.com/   ------------------------------  % Date: Mon, 15 Jan 2001 13:39:40 -0500 - From: JF Mezei <jfmezei.spamnot@videotron.ca>06 Subject: Re: What is the installed base of Tru64 Unix?, Message-ID: <3A6343E7.F492AA4B@videotron.ca>   waybright@my-deja.com wrote: > B > It's well advertised that the installed base for VMS is 450,000+J > systems.  Does anyone have figures on the installed base for Tru64 UNIX?  I I have to seriously doubt that 450,000 VMS istalled base number. How does M Compaq really know if a machine is still being used or not ? Does that numberaE include the cannabalised Microvax II I have in the basement whcih wasxF purchased from a large corporation which migrated from VMS to HP-UX inG Montreal ? (I use it for spare parts, and used the cabinets to store myh0 stereo, and RA drive drawers to store maps etc).  L Or does that 450k number include only those that are still paying support to Compaq ?   ------------------------------   Date: 15 Jan 2001 16:42:28 GMT2 From: mathog@seqaxp.bio.caltech.edu (David Mathog)> Subject: Re: what is the latest version for netscape on VMS..., Message-ID: <93v99k$s3b@gap.cco.caltech.edu>  [ In article <93pmh9$gmi$1@info.service.rug.nl>, helbig@astro.rug.nl (Phillip Helbig) writes:CE >....and where can it be downloaded?  If it is still 3.03, when will i >something newer come along?  : The historical evidence suggests "when hell freezes over".   Sorry,   David Mathog mathog@seqaxp.bio.caltech.edu ? Manager, sequence analysis facility, biology division, Caltech m   ------------------------------  % Date: Mon, 15 Jan 2001 09:23:17 -0500n! From: Beyonder <beyonder@vrx.net>$- Subject: where is the new version of OpenVMS?N8 Message-ID: <qt166tg3ndq5ucmt5gnc5496hc557o1tt3@4ax.com>  4 I have I think 7.2, and there was supposedly a 7.2-1 but where is 7.3 or higher?e   B.   ------------------------------  # Date: Mon, 15 Jan 2001 15:04:58 GMTG= From: system@SendSpamHere.ORG (Brian Schenkenberger, VAXman-)o1 Subject: Re: where is the new version of OpenVMS? 0 Message-ID: <009F6263.14BD5583@SendSpamHere.ORG>  \ In article <qt166tg3ndq5ucmt5gnc5496hc557o1tt3@4ax.com>, Beyonder <beyonder@vrx.net> writes:5 >I have I think 7.2, and there was supposedly a 7.2-1r >but where is 7.3 or higher?   In field test.  G V7.2 and V7.2-1 are the current version for VAX and Alpha respectively.t  U --O VAXman- OpenVMS APE certification number: AAA-0001     VAXman(at)TMESIS(dot)COMr            wO city, n., 1. a place where trees are cut down and streets are named after them.O   ------------------------------  % Date: Mon, 15 Jan 2001 09:30:54 -0600<7 From: "David J. Dachtera" <djesys.nospam@earthlink.net>t1 Subject: Re: where is the new version of OpenVMS?"- Message-ID: <3A6317AE.93C0E0C2@earthlink.net>"   Beyonder wrote:  > 6 > I have I think 7.2, and there was supposedly a 7.2-1 > but where is 7.3 or higher?   . V7.2-1 is current. V7.3 is due out Q1 of 2001.   -- t David J. Dachtera  dba DJE SystemsG http://www.djesys.com/  : Unofficial Affordable OpenVMS Home Page and Message Board: http://www.djesys.com/vms/soho/   F This *IS* an OpenVMS-related newsgroup. So, a certain bias in postings is to be expected.  @ Feel free to exercise your rights of free speech and expression.  F However, attacks against individual posters, or groups of posters, are strongly discouraged.C   ------------------------------  % Date: Mon, 15 Jan 2001 09:51:49 -0500G0 From: David Beatty <David.Beatty.NOSPAM@sas.com>1 Subject: Re: where is the new version of OpenVMS? 2 Message-ID: <Nw5jOnsqExBUMDmjD6HbKrIMZw1m@4ax.com>  @ The latest version is V7.2-1.  There is also a V7.2-1H1 hardware; variant for some of the newer hardware.  V7.3-EFT2 was justO3 released, and should go production later this year..   David R. Beatty_  F On Mon, 15 Jan 2001 09:23:17 -0500, Beyonder <beyonder@vrx.net> wrote:  5 >I have I think 7.2, and there was supposedly a 7.2-1  >but where is 7.3 or higher? >_ >B.i   ------------------------------  % Date: Mon, 15 Jan 2001 09:54:28 -0600 + From: "Main, Kerry" <Kerry.Main@compaq.com>"1 Subject: RE: where is the new version of OpenVMS? N Message-ID: <910612C07BCAD1119AF40000F86AF0D805284C09@kaoexc3.kao.cpqcorp.net>  ; As others have pointed out, VMS V7.2-1 is current release.    C For information on VMS V7.3 that is now in external FT2, reference:   0 <http://www.openvms.compaq.com/openvms/sdk.html>   Regards,  
 Kerry Main Senior ConsultantU Compaq Canada Inc. Professional Servicese Voice: 613-592-4660  Fax  :  819-772-7036 Email: Kerry.Main@Compaq.com     -----Original Message-----( From: Beyonder [mailto:beyonder@vrx.net] Sent: January 15, 2001 9:23 AM To: Info-VAX@Mvb.Saic.Comf- Subject: where is the new version of OpenVMS?h    4 I have I think 7.2, and there was supposedly a 7.2-1 but where is 7.3 or higher?,   B.   ------------------------------  % Date: Mon, 15 Jan 2001 15:33:45 +0000e% From: Alan Greig <agreig@my-deja.com>s1 Subject: Re: where is the new version of OpenVMS?c8 Message-ID: <su566tgt3itbruvvauri3tftdblevst0cj@4ax.com>  F On Mon, 15 Jan 2001 09:23:17 -0500, Beyonder <beyonder@vrx.net> wrote:  5 >I have I think 7.2, and there was supposedly a 7.2-16 >but where is 7.3 or higher? >>    % Currently in external field test. See . http://www.openvms.compaq.com/openvms/sdk.html   Orderable online for $40   --
 Alan Greig   ------------------------------    Date: 15 Jan 2001 13:10:30 -0500, From: koehler@eisner.decus.org (Bob Koehler)1 Subject: Re: where is the new version of OpenVMS? + Message-ID: <0r8LCNdSE2Bf@eisner.decus.org>h  \ In article <qt166tg3ndq5ucmt5gnc5496hc557o1tt3@4ax.com>, Beyonder <beyonder@vrx.net> writes:6 > I have I think 7.2, and there was supposedly a 7.2-1 > but where is 7.3 or higher?  >   D As discussed elsewhere in c.o.v 7.3 is in field test and expected toC ship in the Spring.  Nothing later than that has been announced fore field test yet.a  $ Or were you looking for 9.0 already?  F ----------------------------------------------------------------------? Bob Koehler                     | Computer Sciences Corporation3= NASA GSFC Flight Software       | Federal Sector, Civil Group$E                                 | please remove ".aspm" when replyingw   ------------------------------    Date: 15 Jan 2001 11:07:48 -0500 From: briggs@eisner.decus.org2P Subject: Re: [Change subject] percentage of world's processing on "mainframes" ?+ Message-ID: <FlMaBOrqWLqh@eisner.decus.org>u  K In article <3A622839.F3143C61@Omond.net>, Roy Omond <Roy@Omond.net> writes:  > JF Mezei wrote:lO >> Consider all the CPU power that is used by young (and not so young) kids whoqO >> play all those fancy games with 3D animations, real-time rendering, and manyeP >> networked through the internet. Consider the CPU power use by all those CISCOG >> routers and Sun DNS, NEWS, EMAIL, WEB systems that run the internet.u5 [snip other considerations of MS word and NT servers]t  V > Right, so in a word, "is more than 50% of the total processing today being performed > byE > "mainframes"", then the answer is obviously and unequivocally "NO".   E Well, "NO" is unequivocal.  But not obvious.  I would equivocate that-G it depends on what you mean by "processing" and by "mainframe".  To me,PF network routing, mail service, word processing and game playing aren't6 "processing" and VAXen and Alphas aren't "mainframes".  & 	John Briggs			briggs@eisner.decus.org   ------------------------------   End of INFO-VAX 2001.030 ************************