Stay on Alpha forever? - VMS

This is a discussion on Stay on Alpha forever? - VMS ; In article , moroney@world.std.spaamtrap.com (Michael Moroney) writes: > John Reagan writes: > >>There has been only one case with me where a customer had this rather >>ugly Macro-32 application on Alpha which directly manipulated the FP, >>SP, etc. to emulate ...

+ Reply to Thread
Page 3 of 4 FirstFirst 1 2 3 4 LastLast
Results 41 to 60 of 76

Thread: Stay on Alpha forever?

  1. Re: Stay on Alpha forever?

    In article , moroney@world.std.spaamtrap.com (Michael Moroney) writes:
    > John Reagan writes:
    >
    >>There has been only one case with me where a customer had this rather
    >>ugly Macro-32 application on Alpha which directly manipulated the FP,
    >>SP, etc. to emulate some PDP-11 application. I looked for about 30
    >>minutes and threw up my hands. I recommended they stay on Alpha or go
    >>to a VAX emulator product.

    >
    > (waves!)
    >
    > It's on a VAX, not an Alpha. The Alpha compiler would have choked on the
    > bizarre code just as much as the Itanium compiler does now.


    That is why TECO is emulated on Alpha.

    My one experience with it on Itanium is that it:

    a) Is still emulated
    b) crashes

  2. Re: Stay on Alpha forever?

    On 08/02/07 15:54, Larry Kilgallen wrote:
    > In article , Ron Johnson writes:
    >> On 08/01/07 17:00, Larry Kilgallen wrote:
    >>> In article , moroney@world.std.spaamtrap.com (Michael Moroney) writes:
    >>>
    >>>> Do you need a performance boost now? What kind of performance boost are
    >>>> you likely to need in the future?
    >>> Note that there are some domains where no performance boost is needed.
    >>> The best example I have heard of is the software reading toll tickets
    >>> on toll roads. That is a case where the bandwidth of the road is never
    >>> going to increase enough to tax the computer capacity. And I am quite
    >>> careful before I use the word "never".

    >> More plazas, more lanes, especially more customers, and, if various

    >
    > No, the bandwidth limit is the road.


    A fully-saturated road where everyone uses a toll tag is, obviously,
    an upper limit on usage.

    However, the roads are not saturated, and not everyone uses toll tags.

    (Toll roads and VMS are my job.)

    >> agencies have sharing agreements (think E-ZPass, for those of you in
    >> the NE United States), more reciprocity processing.
    >>
    >> Eventually, "more, more, more" maxes out the box.

    >
    > Do you have an example of this ? The one discussed at DECUS for many
    > years switched to Alpha from PDP-11 due to parts availability, not for
    > performance reasons.


    Yes. One of our toll systems is upgrading from a GS160 to a GS1280
    because it's over-burdened.

    But it's not a "lane computer", simply processing tags as they zip
    by. 80486 computers running Linux can do that (including taking 2
    pictures of every car).

    --
    Ron Johnson, Jr.
    Jefferson LA USA

    Give a man a fish, and he eats for a day.
    Hit him with a fish, and he goes away for good!

  3. Re: Stay on Alpha forever?

    On 08/02/07 15:22, John Wallace wrote:
    [snip lots of agreement]
    > unless it's ECC memory. ECC still isn't common on desktop/laptop machines,
    > but 1GB or more of memory is not unusual these days, which makes me wonder
    > how many crashes folks are going to see because of undetected main memory
    > errors. They'll mostly just blame Windows of course, and historically they'd
    > have been right to do so.


    Anecdote:

    I've had 1GB of RAM on my desktop PC for ~4 years now (it's an
    always-on system running Linux, so don't think that it's only
    running 3 hours/day or some such), and have never had memory "issues".

    --
    Ron Johnson, Jr.
    Jefferson LA USA

    Give a man a fish, and he eats for a day.
    Hit him with a fish, and he goes away for good!

  4. Re: Stay on Alpha forever?

    On Aug 2, 4:22 pm, "John Wallace"
    wrote:
    > "Neil Rieck" wrote in message
    >

    [...snip...]
    >
    > Please don't compare chalk and cheese. If you wish, compare current Itaniums
    > with something like similarly priced Proliant-class boxes, not with el
    > cheapo commodity-priced desktop-based stuff. Compaq's better Proliants have
    > had plenty of RAS features for many years, and the research I have done says
    > they (and their equivalents from IBM, Sun, etc) compare quite reasonably
    > with Itanium boxes, at least at the low end.
    >


    I couldn't agree more. I've got lots of Proliants instlled in the back
    room, many still running Windows-NT4 (don't ask) and they are
    reasonably stable and only have a problem every 12 months on average.
    So when the Proliant maintenance guy comes in to swap boards on one of
    them I always point out the AS-2100 running beside them; and it blows
    his mind when I show him that the uptime display indicates this thing
    has been serving up DHCP addresses without a reboot since July-2004.

    >
    > Also, you might want to refer to ECC memory rather than parity memory;
    >

    Thanks for correcting me. I remember the ECC Lecture part of the
    PDP-11/44 course at DEC in Kanata (Ottawa). 32 bits of data and 5 bits
    of protection. Even back then the PC people refused to accknowledge
    that this stuff was necessary.

    Neil Rieck
    Kitchener/Waterloo/Cambridge,
    Ontario, Canada.
    http://www3.sympatico.ca/n.rieck/


  5. Re: Stay on Alpha forever?

    In article <1185995866.368720.29970@g4g2000hsf.googlegroups.co m>, tadamsmar writes:
    >
    > Can a costly upgrade be justfied simply because you can't stay on
    > Alphas forever?


    We still have mission-critical systems on VAXen. No problem, we
    can get parts.

    I've recently had to get parts for an 11/785, not terribly hard.
    I used to have problems getting parts for six month old SGI.

    You mileage will vary. Talk to your hardware maintenance folks
    about parts availability for the Alphas you have.


  6. Re: Stay on Alpha forever?

    In article <65d84$46b132bf$cef8887a$22210@TEKSAVVY.COM>, JF Mezei writes:
    >
    > As long as new versions of VMS-Alpha are being generated


    That's not what many mission critical apps need. None of my mission
    critical VAXen run anything later than 6.2.


  7. Re: Stay on Alpha forever?

    In article <1186050866.271931.120540@i13g2000prf.googlegroups. com>, Neil Rieck writes:
    >
    > I did two ports from VAX to Alpha between 1999-2001. The VAX-BASIC
    > stuff was straight forward although I decided to rewrite some code
    > produced by some less disciplined programmers; I was slowed down a
    > little more by converting "some highly tweaked VAX-C code to DEC-C"
    > and "Macro-32 code to DEC-C". Now that the hard work is done I expect
    > a port from Alpha to Itanium should be fairly easy.


    All my code, invluding Macro-32, just recompiled in going from VAX to
    Alpha. The only problems I encountered were limitations of the
    Fortran compiler which shipped with OpenVMS Alpha AXP 1.5. Those
    went away when I upgraded to 6.0 and the later compiler.

    Porting other folks Macro-32 has simply been a matter of reading and
    following the porting guide, adding some new directives that have
    long since been accepted by the VAX version of Macro-32.


  8. Re: Stay on Alpha forever?

    In article , "Tom Linden"
    writes:

    > I found that Alpha executables are about 3x those of VAX,


    This is mainly RISC as opposed to CISC. Even if CISC is inherently
    better, the longer time between design and fabrication means that, as
    long as Moore's law and current times to market don't change much, it
    will always lag behind RISC in terms of performance. Of course, memory
    is so cheap these days that this cannot really be a problem for anybody.

  9. Re: Stay on Alpha forever?

    On Fri, 03 Aug 2007 08:26:19 -0700, Phillip Helbig---remove CLOTHES to
    reply wrote:

    > In article , "Tom Linden"
    > writes:
    >
    >> I found that Alpha executables are about 3x those of VAX,

    >
    > This is mainly RISC as opposed to CISC. Even if CISC is inherently
    > better, the longer time between design and fabrication means that,

    That is an unproven assertion.
    as
    > long as Moore's law and current times to market don't change much, it

    It isn't a law, of course, it is an observation.

    > will always lag behind RISC in terms of performance.

    I don't buy that.

    Of course, memory
    > is so cheap these days that this cannot really be a problem for anybody.





    --
    PL/I for OpenVMS
    www.kednos.com

  10. Re: Stay on Alpha forever?

    Richard B. Gilbert wrote:
    > I've no personal experience but I've read that Itanic has little or no
    > performance advantage over the best Alphas.


    This is no longer the case. From Greg Jordan's "OpenVMS Performance
    Update" session at HP Technology Forum 2007:

    "The current Integrity systems perform better than existing Alpha
    systems in most cases
    - often by substantial amounts, and with:
    o lower acquisition costs
    o reduced floor and rack space requirements
    o reduced cooling requirements"

  11. Re: Stay on Alpha forever?

    On Fri, 03 Aug 2007 09:33:24 -0700, Keith Parris
    wrote:

    > Richard B. Gilbert wrote:
    >> I've no personal experience but I've read that Itanic has little or no
    >> performance advantage over the best Alphas.

    >
    > This is no longer the case. From Greg Jordan's "OpenVMS Performance
    > Update" session at HP Technology Forum 2007:
    >
    > "The current Integrity systems perform better than existing Alpha
    > systems in most cases
    > - often by substantial amounts, and with:
    > o lower acquisition costs
    > o reduced floor and rack space requirements
    > o reduced cooling requirements"


    Of course, performance isn't everything. Capability is far more important.


    --
    PL/I for OpenVMS
    www.kednos.com

  12. Re: Stay on Alpha forever?

    Michael Moroney wrote:
    > Will the Itanic upgrade pay for itself in reduced maint. cost or power
    > consumption (wait...Itanic power consumption...scratch that argument!)?


    While some of the earlier Itanium chips were high-wattage, in the latest
    generation of chips Intel seems to have improved the process to have
    much lower leakage current, so even with their higher performance the
    power consumption is lower than the older chips.

  13. Re: Stay on Alpha forever?

    etmsreec@yahoo.co.uk wrote:
    > Memory bandwidth isn't (I'm told) as good as it was on Alpha. I think
    > that's a "yet" rather than a "will never be".


    Greg Jordan's OpenVMS Performance Update session from HPTF 2007 shows
    the rx7640 and rx8640 have higher memory bandwidth than the GS-1280, and
    the rx2600, rx2620, rx3600, and rx4640 have higher memory bandwidth than
    the ES45 (the rx3600 actually has about twice the memory bandwidth of
    the ES45).

  14. Re: Stay on Alpha forever?

    Larry Kilgallen wrote:
    > You don't get much higher level that Ada or PL/I, the two notable
    > discontinuities where HP has declined to provide support.


    Be sure to correctly understand "HP has declined to provide support". In
    particular, is *not* the same as "no support."

    Ada is available on OpenVMS I64 from a third party, AdaCore. Customers
    can obtain support for this compiler from AdaCore if they wish.

    With respect to PL/I, I know of no solution available on OpenVMS I64,
    whether from HP or from anyone else. I don't claim to be an expert in
    PL/I compiler availability, so my knowledge should be considered incomplete.

    -Jeff

  15. Re: Stay on Alpha forever?

    On Fri, 03 Aug 2007 10:57:08 -0700, Keith Parris
    wrote:

    > etmsreec@yahoo.co.uk wrote:
    >> Memory bandwidth isn't (I'm told) as good as it was on Alpha. I think
    >> that's a "yet" rather than a "will never be".

    >
    > Greg Jordan's OpenVMS Performance Update session from HPTF 2007 shows
    > the rx7640 and rx8640 have higher memory bandwidth than the GS-1280, and
    > the rx2600, rx2620, rx3600, and rx4640 have higher memory bandwidth than
    > the ES45 (the rx3600 actually has about twice the memory bandwidth of
    > the ES45).

    Yes, but it needs higher bandwidth just to offer the same throughput owing
    to the EPIC architecture


    --
    PL/I for OpenVMS
    www.kednos.com

  16. Re: Stay on Alpha forever?

    Keith Parris wrote:

    > "The current Integrity systems perform better than existing Alpha
    > systems in most cases


    It is a tribute to Alpha that a 2001 generation chip (that is when EV7
    was originally expected, right ?) still prevents a 2007 IA64 from
    claiming "in all cases".



  17. Re: Stay on Alpha forever?

    Keith Parris wrote:
    > Greg Jordan's OpenVMS Performance Update session from HPTF 2007 shows
    > the rx7640 and rx8640 have higher memory bandwidth than the GS-1280, and
    > the rx2600, rx2620, rx3600, and rx4640 have higher memory bandwidth than
    > the ES45 (the rx3600 actually has about twice the memory bandwidth of
    > the ES45).


    Sorry for my memory failures (no EEC or Parity in my bio memory pack),
    but is VMS qalified to sun on Superdomes yet ? Last I heard, they made
    it run on those C class wintel blade server enclosures that also support
    IA64 cards. Don't recall hearing about VMS on superdomes.

    If not yet available on superdomes, I am curious about why, what sort of
    technical architecture would prevent VMS from taking advantage of a
    Superdome. Or is it purely a political decision to not allow VMS to take
    better advantage of a superdome vs HP-UX ?

  18. Re: Stay on Alpha forever?

    Jeff wrote:
    > Larry Kilgallen wrote:
    > Be sure to correctly understand "HP has declined to provide support". In
    > particular, is *not* the same as "no support."


    In the case of PL1, the story has been alluded to here on cov some time
    ago. On Alpha, the PL1 compiler has full access to te GEM stuff. HP has
    declined to provide access to the GEM stuff on IA64 which would have
    allowed a proper port of the product.

    I would use "refused" intead of "declined" since the decision to prevent
    the company to port PL1 was a very active decision.

  19. Re: Stay on Alpha forever?

    On 2 Aug 2007 at 20:22, John Wallace wrote:
    > By far the least reliable bit of a typical Proliant these days is not
    > hardware, it's Windows. But try telling that to the people doing
    > VMware-based (or similar) server consolidation (or to those about to
    > throw away a real VMS box to replace it with Charon?).


    I hate Windows, too. But it's an okay environment to run a singe
    application. I used to develop a product that ran on DOS -- it was
    great as long as you didn't expect much.

    As far as CHARON goes, it IS a VMS box. And 90% of my customers want
    Windows. Not for technical reasons, mind you. Don't complain to me -
    -it's just the way it is.

    You want CHARON on Linux (or some other OS)? Commit to buy a whole
    bunch of copies, and I can get that done.
    --Stan Quayle
    Quayle Consulting Inc.

    ----------
    Stanley F. Quayle, P.E. N8SQ Toll free: 1-888-I-LUV-VAX
    8572 North Spring Ct., Pickerington, OH 43147 USA
    stan-at-stanq-dot-com http://www.stanq.com/charon-vax.html
    "OpenVMS, when downtime is not an option"



  20. Re: Stay on Alpha forever?

    JF Mezei wrote:
    > (Buying it from HP probably requires you know some magic incantation,
    > and/or go through Cerner which supposedly has a deal to continue to make
    > Alphas available to its clients)


    No magic incantation needed: see http://www.hp.com/products/hprenew/

+ Reply to Thread
Page 3 of 4 FirstFirst 1 2 3 4 LastLast