1 INFO-VAX	Sat, 22 Jul 2006	Volume 2006 : Issue 404       Contents: Re: Alpha remembrance day  Re: Alpha remembrance day ) Re: New itaniums out at 2.5x perform gain ( Re: Tomcat user authentication question.A Re: Under VMS, on an HSG80, can Raid Partition Size be Increased?   F ----------------------------------------------------------------------  % Date: Fri, 21 Jul 2006 13:53:48 -0400 ( From: Bill Todd <billtodd@metrocast.net>" Subject: Re: Alpha remembrance dayG Message-ID: <WsCdnRKkwfswiVzZnZ2dnUVZ_t2dnZ2d@metrocastcablevision.com>   
 Andrew wrote:    ...   1   The decisions to design and build Alpha and the G > decisions about what OS platforms it would support and what migration G > options would be offered to existing customers had already been made. G > The only thing Palmer could have done at that point was cancelled the 
 > project.  G You still don't seem able to wrap what passes for your mind around the  D fact that a product cannot possibly *decline* due to decisions made F public before it was even *born* (i.e., before it was released to the H market to establish an actual position - in the context of those public 6 positions - from which a *decline* could be measured).   > I > As a perspective on Alpha from a rather better informed source than you H > read the comments made by Michael Slater in an interview in 1992 after > Alpha's announcement.  > G > UPSIDE: We didn't talk much about the Alpha chip. Is it going to be a  > world-beater?  > H > SLATER: I don't think so. Alpha is an outstanding piece of technology.G > It is probably superior architecture to the architectures that exist, I > but I don't believe the differences are big enough to overcome the fact G > that it is very late entering the market. There is an enormous amount H > of inertia and software base that DEC has to fight. Every company thatG > DEC goes out to convince to make Alpha chips or to build systems with F > Alpha, every one of those companies over the last few years has beenF > pitched by Sun for Sparc, pitched by Mips, then HP for PA RISC, then > pitched by IBM for RS/6000.  > E > How precient of him the reasons for the death of Alpha spelt out in  > 1992 just after its launch.   E In your own little alternative reality, perhaps.  In the real world,  H Alpha and Tru64 indeed struggled to reestablish credibility in the Unix D market, but the struggle (as I've noted before - but which you keep G ignoring because it proves inconvenient to your myth) was an *upward*,  G not a *downward*, struggle.  Tru64 was growing robustly during the two  H years that preceded the Alphacide (and possibly earlier, though I don't E happen to have information about that period) and had already clawed  E back a respectable chunk of the market; VMS was growing (albeit much  G more slowly) during a similar period; and the combination of Alpha and  D its prime OSs was one of the most profitable franchises (along with H service and storage) that Compaq owned - *far* more profitable than its - faltering (at least profit-wise) PC business.   E Deal with that fact, or shut up:  there are many reasons why Alpha's  F growth was not *more robust than it was*, and we even appear to agree H about some of them, but  a) any lack of applications was at most one of F many such reasons (rather than some kind of critical flaw that all by I itself could have prevented Alpha from more robust growth) and  b) there  / was no actual *decline* prior to the Alphacide.    - bill   ------------------------------  % Date: Fri, 21 Jul 2006 16:09:57 -0400 - From: JF Mezei <jfmezei.spamnot@teksavvy.com> " Subject: Re: Alpha remembrance day, Message-ID: <44C13489.98470EF1@teksavvy.com>  
 Andrew wrote: G > What a truly ludicrous point. By the time Palmer came on the scene in H > 1992 Alpha had been a fully funded project since 1989 and was itself aF > followon from PRISM. The decisions to design and build Alpha and theG > decisions about what OS platforms it would support and what migration G > options would be offered to existing customers had already been made. G > The only thing Palmer could have done at that point was cancelled the 
 > project.  B It was under Palmer that Digital's software portfolio was severelyG reduced and that many products abanmdonned/not ported to Alpha (FMS was H one of them, a decisions later reversed). Rememeber also the dropping ofH VAX document/bookreader in favour of some unknown 3rd party, a decisionsF that was also rescinded much later after damage had already been done.    H It was under Palmer that decisions were made to refuse to use the HudsonE fab to its fullest, reserving production capacity for Alpha should it H ever take off, this, at the same time that Palmer made damned sure Alpha/ couldn't grow to compete angainst Intel's 8086.   D It was also Palmer that killed one of the biggest profit centres forF DEC: ALL-IN-1. Shortly after they ad announced the porting of ALL-IN-1D to Unix and Windows, Palmer announced that he had struck a deal withF Microsoft to deploy Office everywhere and abandon ALL-IN-1 and all the= messaging infrastructuire that had been so profitable to DEC.   H Prior to Palmer, the technical decisions/development of Alpha were made.F Under Palmer, it was the business decisions that were made about Alpha7 and it is those that crippled Alpha and killed Digital.       H > SLATER: I don't think so. Alpha is an outstanding piece of technology.G > It is probably superior architecture to the architectures that exist, I > but I don't believe the differences are big enough to overcome the fact + > that it is very late entering the market.   G Ironic statement in hindsight isn't it ? It was in fact extremely early B into the market. But His point still has validity in that the 8086* architecture had already spread its roots.  F However, there would have still be chance for Alpha to uproot the 8086E in the 1990s, especially during the hectic "fleet replacement" cycles G for Win95/NT and later the Y2K. But by then, Plamer had already decided - to not allow Alpha to compete at the low end.    ------------------------------    Date: 21 Jul 2006 11:31:23 -0700 From: perfnerd@yahoo.com2 Subject: Re: New itaniums out at 2.5x perform gainB Message-ID: <1153506683.600694.140020@i3g2000cwc.googlegroups.com>   Bill Todd wrote: > perfnerd@yahoo.com wrote:  >  > ...  > 1 >   SPEC is just to well understood and tuned for K > > increases in cache and memory speed to show a significant difference in  > > performance. > G > Hmmm.  While I don't have time to check right now, my recollection is F > that Madison II submissions differing only in cache size (same clock@ > speed, same hardware, same software, and I think even the sameJ > submission date) achieved noticeably (though not dramatically) different > SPECint scores.   D I think that is the 3/6/9 meg cache versions.  If you look at the 3MF system, it actually out performs the 6M and 9M CPUs, which I attributeF to it having a faster memory bus.  So, I do know that increasing cacheE size and/or memory speed does have an effect on certain components of E SPEC.  It just looks like they are getting close to 'enough' cache as G far as most of the SPEC components are concerned.  The numbers are just G not showing a linear performance improvement coming from a 33% increase F in L3 cache and 33% increase in memory bus speed (with faster memory).C That would imply, to me anyway, that more often than not the CPU is D waiting on data already in the cache and adding cache only increased  the hit ratio by a small amount.   ------------------------------  % Date: Fri, 21 Jul 2006 21:29:55 -0400 7 From: =?ISO-8859-15?Q?Arne_Vajh=F8j?= <arne@vajhoej.dk> 1 Subject: Re: Tomcat user authentication question. . Message-ID: <jcfwg.70835$fG3.17528@dukeread09>   Tom Linden wrote: H > On Wed, 19 Jul 2006 14:04:46 -0700, Rick Barry <richard.barry@hp.com>  > wrote:E >> CSWS_JAVA (Tomcat) does not yet contain a built-in authentication   >> mechanismL >> for OpenVMS. You could architect such a mechanism using JAAS and Tomcat'sJ >> JAAS Realm plug-in, but you'll need to do a bit of coding using the JNIF >> interface. We may add this to a future version of CSWS_JAVA. In the' >> meantime, you need to roll your own.  > < > Alternatively, you use a web server designed for VMS, WASD  ; Considering that Tomcat is *not* a web server but a servlet 5 container, then I doubt that WASD would do much good.    Arne   ------------------------------    Date: 21 Jul 2006 11:08:14 -0700' From: "syslost" <wm.reynolds@gmail.com> J Subject: Re: Under VMS, on an HSG80, can Raid Partition Size be Increased?B Message-ID: <1153505294.018088.22900@m79g2000cwm.googlegroups.com>  D I was planning on doing a backup/restore, but thought increasing theC size of the partition would be faster.  I couldn't find anything on E increasing partition sizes in the hsg80 documentation and wondered if 3 there were other products or undocumented features.      The KGB wrote:G > I didn't see any replies so here's my 2 centz!  Once it's set you can L > INcrease it but not DEcrease the size of the RAID.  Somebody may know of aL > commercial product that will accomodate you but why not just copy the data) > off, create what you need then restore? 4 > "syslost" <wm.reynolds@gmail.com> wrote in message? > news:1153327789.196166.104250@s13g2000cwa.googlegroups.com...    ------------------------------   End of INFO-VAX 2006.404 ************************