The Common System Interface: Intel's Future Interconnect - VMS

This is a discussion on The Common System Interface: Intel's Future Interconnect - VMS ; All, The following article may be of interest: (August 28, 2007) http://www.realworldtech.com/page.cf...2807020032&p=1 "In the competitive x86 microprocessor market, there are always swings and shifts based on the latest introductions from the two main protagonists: Intel and AMD. Thenext anticipated shift ...

+ Reply to Thread
Page 1 of 2 1 2 LastLast
Results 1 to 20 of 29

Thread: The Common System Interface: Intel's Future Interconnect

  1. The Common System Interface: Intel's Future Interconnect

    All,

    The following article may be of interest: (August 28, 2007)
    http://www.realworldtech.com/page.cf...2807020032&p=1
    "In the competitive x86 microprocessor market, there are always swings and shifts based on
    the latest introductions from the two main protagonists: Intel and AMD. Thenext anticipated
    shift is coming in 2008-9 when Intel will finally replace their front side bus architecture.
    This report details Intel's next generation system interconnect and the associated cache
    coherency protocol, likely deployment plans across the desktop, notebook and server market
    as well as the economic implications."

    [Snip - see article]

    Regards


    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.




  2. Re: The Common System Interface: Intel's Future Interconnect

    Main, Kerry wrote:
    > All,
    >
    > The following article may be of interest: (August 28, 2007)
    > http://www.realworldtech.com/page.cf...2807020032&p=1



    Mr Main,

    What will differentiate a 64 bit 8086 plugged into a CSI interface from
    a 64 bit IA64 also plugged into a CSI interface ?

    If both have access to the same type of memory, cache etc, then won't
    the industry standard architecture that has competition from AMD end up
    being far superior than some proprietary IA64 thing that requires its
    own proprietary funky compilers due to ist EPIC nature ?

  3. RE: The Common System Interface: Intel's Future Interconnect


    > -----Original Message-----
    > From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca]
    > Sent: August 29, 2007 9:35 PM
    > To: Info-VAX@Mvb.Saic.Com
    > Subject: Re: The Common System Interface: Intel's Future Interconnect
    >
    > Main, Kerry wrote:
    > > All,
    > >
    > > The following article may be of interest: (August 28, 2007)
    > > http://www.realworldtech.com/page.cf...2807020032&p=1

    >
    >
    > Mr Main,
    >
    > What will differentiate a 64 bit 8086 plugged into a CSI interface from
    > a 64 bit IA64 also plugged into a CSI interface ?
    >
    > If both have access to the same type of memory, cache etc, then won't
    > the industry standard architecture that has competition from AMD end up
    > being far superior than some proprietary IA64 thing that requires its
    > own proprietary funky compilers due to ist EPIC nature ?


    JF -

    Remember the relative importance to Cust's in terms of both what they have and what they
    need In the future:

    1. App = 50-60%
    2. OS = 25-35%
    3. Server HW - 10-15%

    While #3 gets all sorts of attention in techie newsgroups, in the pig picture,
    #1 and #2 are much more important to Cust's. With a massive glut in available compute
    cycles in most Cust's environment today, Cust's are not impressed with fantastic new
    computer speeds that will increase their glut of available compute cycles even more.

    Btw, this applies to all platforms.

    Also, keep in mind that there is now a massive trend to consolidating both servers and
    DC's. This is a huge change from the distributed computing designs of the last 10 years.

    Imho, the question that will become increasingly important in the future - "Can a company
    afford OS platforms for their future centralized, very HA strategy that have "one app,
    one OS" App/ISV support cultures and where the OS vendors release 5-20 security patches
    per month?"

    My advice to Cust's - think like Wayne Gretzky (ok, he's a hockey player) and the way
    He became a great player .. "Do not skate to where the puck is, but where it will be in
    the next play sequence .."

    :-)

    Regards


    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.






  4. Re: The Common System Interface: Intel's Future Interconnect

    In article ,
    "Main, Kerry" writes:
    >
    >> -----Original Message-----
    >> From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca]
    >> Sent: August 29, 2007 9:35 PM
    >> To: Info-VAX@Mvb.Saic.Com
    >> Subject: Re: The Common System Interface: Intel's Future Interconnect
    >>
    >> Main, Kerry wrote:
    >> > All,
    >> >
    >> > The following article may be of interest: (August 28, 2007)
    >> > http://www.realworldtech.com/page.cf...07020032&p=3D1

    >>
    >>
    >> Mr Main,
    >>
    >> What will differentiate a 64 bit 8086 plugged into a CSI interface from
    >> a 64 bit IA64 also plugged into a CSI interface ?
    >>
    >> If both have access to the same type of memory, cache etc, then won't
    >> the industry standard architecture that has competition from AMD end up
    >> being far superior than some proprietary IA64 thing that requires its
    >> own proprietary funky compilers due to ist EPIC nature ?

    >
    > JF -
    >
    > Remember the relative importance to Cust's in terms of both what they have =
    > and what they
    > need In the future:
    >
    > 1. App =3D 50-60%
    > 2. OS =3D 25-35%
    > 3. Server HW - 10-15%
    >
    > While #3 gets all sorts of attention in techie newsgroups, in the pig pictu=
    > re,
    > #1 and #2 are much more important to Cust's. With a massive glut in availab=
    > le compute
    > cycles in most Cust's environment today, Cust's are not impressed with fant=
    > astic new
    > computer speeds that will increase their glut of available compute cycles e=
    > ven more.
    >
    > Btw, this applies to all platforms.
    >
    > Also, keep in mind that there is now a massive trend to consolidating both =
    > servers and
    > DC's. This is a huge change from the distributed computing designs of the l=
    > ast 10 years.
    >
    > Imho, the question that will become increasingly important in the future - =
    > "Can a company
    > afford OS platforms for their future centralized, very HA strategy that hav=
    > e "one app,
    > one OS" App/ISV support cultures and where the OS vendors release 5-20 secu=
    > rity patches
    > per month?"
    >
    > My advice to Cust's - think like Wayne Gretzky (ok, he's a hockey player) a=
    > nd the way
    > He became a great player .. "Do not skate to where the puck is, but where i=
    > t will be in
    > the next play sequence .."
    >
    >:-)


    Kerry,
    I don't mean to be obtuse but not only did you not answer any of the
    questions in the message you were replying to but one could easily draw
    the conclusion that if apps make up 50-60% of the reason for choosing a
    platform your evidence actually supports JF's implications about which
    platform has the better chance for success!!

    bill

    --
    Bill Gunshannon | de-moc-ra-cy (di mok' ra see) n. Three wolves
    bill@cs.scranton.edu | and a sheep voting on what's for dinner.
    University of Scranton |
    Scranton, Pennsylvania | #include

  5. Re: The Common System Interface: Intel's Future Interconnect

    On 08/29/07 20:35, JF Mezei wrote:
    > Main, Kerry wrote:
    >> All,
    >>
    >> The following article may be of interest: (August 28, 2007)
    >> http://www.realworldtech.com/page.cf...2807020032&p=1

    >
    >
    > Mr Main,
    >
    > What will differentiate a 64 bit 8086 plugged into a CSI interface from
    > a 64 bit IA64 also plugged into a CSI interface ?
    >
    > If both have access to the same type of memory, cache etc, then won't
    > the industry standard architecture that has competition from AMD end up
    > being far superior than some proprietary IA64 thing that requires its
    > own proprietary funky compilers due to ist EPIC nature ?


    Ssshhh! Causing Kerry's head to explode from the paradox would
    upset Mrs. Main.

    --
    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!

  6. RE: The Common System Interface: Intel's Future Interconnect

    > -----Original Message-----
    > From: bill@triangle.cs.uofs.edu [mailto:bill@triangle.cs.uofs.edu] On
    > Behalf Of Bill Gunshannon
    > Sent: August 30, 2007 9:24 AM
    > To: Info-VAX@Mvb.Saic.Com
    > Subject: Re: The Common System Interface: Intel's Future Interconnect
    >
    >
    > Kerry,
    > I don't mean to be obtuse but not only did you not answer any of the
    > questions in the message you were replying to but one could easily draw
    > the conclusion that if apps make up 50-60% of the reason for choosing a
    > platform your evidence actually supports JF's implications about which
    > platform has the better chance for success!!
    >
    > bill
    >


    JF asked why a Cust might choose an Integrity based system with CSI over a similar
    X86 system - I assume he meant CSI or AMD based. The answer which I provided, but
    which you appear to not believe, is that the App and OS play much greater into
    the overall equation than low level HW.

    The CSI consideration is even lower on the ladder that the CPU platform decision.

    My point is that Cust's no longer make major decisions based on low level hardware bits
    and bytes. And the wiz bang speed stuff no longer impresses them as it usedto. Cust's
    already have a huge and extremely embarrassing glut of available CPU cyclesin their
    environment.

    What will get their attention is being able to run secure, very HA environments in
    much more centralized environments in a more cost effective manner with much fewer
    resources. Hardware costs are a very small part of the overall equation.

    Hence, the OS and application criteria will be the bigger decision that hasgreater
    weight than whether Integrity-CSI is x% faster than X86-CSI/AMD or vice versa.

    If a Cust thinks Windows/Linux is good enough for a particular one app, oneOS
    application environment and they can deal with all of the monthly security patches,
    then they can choose either the x86 CSI or the AMD system or the Integrity CSI
    system. That's their decision.

    If the Cust values higher security, higher HA (local and multi-site), app stacking,
    native common file system and print/batch queues across all servers on all sites etc,
    then they can choose OpenVMS and Integrity.

    Regards


    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.






  7. Re: The Common System Interface: Intel's Future Interconnect

    Main, Kerry wrote:
    > JF asked why a Cust might choose an Integrity based system with CSI over a similar
    > X86 system - I assume he meant CSI or AMD based. The answer which I provided, but
    > which you appear to not believe, is that the App and OS play much greater into
    > the overall equation than low level HW.


    You know perfectly well what my question was. And the fact that you
    chose to avoid the response means that you know very well what the
    response is and know that it is not a response that an HP employee can
    make in a public forum: AKA: IA64 will not longer have a large enough
    niche left to warrant it continued existance since 8086 will scale even
    further upwards.

    The 8086 is the architecture that has all the activity, the OS, the
    development, and most importantly, the applications. In cases where you
    have an architecture that clearly outperforms the 8086, then you may be
    able to justify porting to that architecture to benefit from the added
    performance. But in cases where a different architecture has similar or
    just marginally better performance, it just isn't worth even considering
    a port to it.


    For as long as HP refuses to acknowledge that IA64 has a bad image and
    uncertain future, for as long as HP refuses to commit to porting VMS
    beyond IA64, then how can one be taken seriously by trying to get new
    customers onto VMS ?

    The fact that HP has decided to tell Cerner to drop VMS in favour of
    HP-UX is a HUGE deal in my opinion. It is basically confirmation that HP
    has no intentions of continuing VMS development once the IA64
    architecture is declared mature.

    Yes, the " is killing VMS" cries have been made for the last
    decade and a half. But this time, I fear that this is it. The end of the
    line. Except HP doesn't have the guts to actually say so, they will
    just announce that VMS is still actively being developped even when
    they are down to just one indian teenager minding the source code from
    his bamboo shack on a beach.


  8. RE: The Common System Interface: Intel's Future Interconnect

    > -----Original Message-----
    > From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca]
    > Sent: August 30, 2007 12:27 PM
    > To: Info-VAX@Mvb.Saic.Com
    > Subject: Re: The Common System Interface: Intel's Future Interconnect
    >
    > Main, Kerry wrote:
    > > JF asked why a Cust might choose an Integrity based system with CSI

    > over a similar
    > > X86 system - I assume he meant CSI or AMD based. The answer which I

    > provided, but
    > > which you appear to not believe, is that the App and OS play much

    > greater into
    > > the overall equation than low level HW.

    >
    > You know perfectly well what my question was. And the fact that you
    > chose to avoid the response means that you know very well what the
    > response is and know that it is not a response that an HP employee can
    > make in a public forum: AKA: IA64 will not longer have a large enough
    > niche left to warrant it continued existance since 8086 will scale even
    > further upwards.
    >


    I answered twice.

    Its not the server HW that primarily matters, but what is running on it.

    You did not read my response as you viewed this as simply another opportunity to flog
    your personal views that the primarily "One App, One OS" x86 server HW platform is going
    to take over the server world like it did the desktop.

    I disagree. I think the HW is becoming an increasingly small factor in IT server decisions,
    especially with such massive consolidation efforts going on right now (70-80% of which
    is X86 focussed).

    And btw, if I quit HP right now, my response tomorrow would be exactly the same.

    However, feel free to continue to promote that 8086 thingy.

    :-)

    Regards


    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.






  9. Re: The Common System Interface: Intel's Future Interconnect

    Main, Kerry wrote:
    > You did not read my response as you viewed this as simply another opportunity to flog
    > your personal views that the primarily "One App, One OS" x86 server HW platform is going
    > to take over the server world like it did the desktop.



    8086 is not a "one app one server platform". Windows is. You can have a
    serious 64 bit 8086 server running a serious operating system which
    supports multiple applications.


    The point is that the 8086 is coming of age and with common interfaces
    with IA64, it will become possible to build large systems that have
    similar or better performance with an 8086 compared to IA64. At that
    point, there is no point in continuing to try to shove an unpopular
    platform down customer's throats.

    There is no reason HP couldn't build a CSI based 8086 with EFI console
    with all the same bells and whistles as IA64 systems. And once this
    happens, there is also no reason to continue to build IA64 based systems.

  10. RE: The Common System Interface: Intel's Future Interconnect

    In article , "Main, Kerry" writes:
    >>I disagree. I think the HW is becoming an increasingly small factor in IT

    >server decisions,


    Yup. As long it is x86 (because the software only runs there now)...

    --
    Peter "EPLAN" LANGSTOEGER
    Network and OpenVMS system specialist
    E-mail peter@langstoeger.at
    A-1030 VIENNA AUSTRIA I'm not a pessimist, I'm a realist

  11. Re: The Common System Interface: Intel's Future Interconnect

    On Aug 30, 8:05 am, "Main, Kerry" wrote:
    > > -----Original Message-----
    > > From: JF Mezei [mailto:jfmezei.spam...@vaxination.ca]
    > > Sent: August 29, 2007 9:35 PM
    > > To: Info-...@Mvb.Saic.Com
    > > Subject: Re: The Common System Interface: Intel's Future Interconnect

    >
    > > Main, Kerry wrote:
    > > > All,

    >
    > > > The following article may be of interest: (August 28, 2007)
    > > >http://www.realworldtech.com/page.cf...2807020032&p=1

    >
    > > Mr Main,

    >
    > > What will differentiate a 64 bit 8086 plugged into a CSI interface from
    > > a 64 bit IA64 also plugged into a CSI interface ?

    >
    > > If both have access to the same type of memory, cache etc, then won't
    > > the industry standard architecture that has competition from AMD end up
    > > being far superior than some proprietary IA64 thing that requires its
    > > own proprietary funky compilers due to ist EPIC nature ?

    >
    > JF -
    >
    > Remember the relative importance to Cust's in terms of both what they have and what they
    > need In the future:
    >
    > 1. App = 50-60%
    > 2. OS = 25-35%
    > 3. Server HW - 10-15%


    I really didn't want to get into this again, but let's talk about
    those numbers. You say "relative importance to Cust's" of the App is
    50-60% ?? I say relative importance of the App is 100% and the OS &
    hardware (and I include workstations, printers and all other
    peripherals and infrastructure) fall into the category of "necessary
    evils." Your numbers more closely reflect the rough cost of acquiring
    the ability to perform the application.

    Platform security, support, reliability, performance and all other OS/
    hardware factors effect cost and are certainly important but they are
    subordinate to and dependent upon the needs of the application.

    You don't buy an application you can't run, and you don't buy a
    platform that won't run your application. Before anyone else says it,
    I'll say it first: "Duh!"


    >
    > While #3 gets all sorts of attention in techie newsgroups, in the pig picture,
    > #1 and #2 are much more important to Cust's. With a massive glut in available compute
    > cycles in most Cust's environment today, Cust's are not impressed with fantastic new
    > computer speeds that will increase their glut of available compute cycles even more.
    >
    > Btw, this applies to all platforms.
    >
    > Also, keep in mind that there is now a massive trend to consolidating both servers and
    > DC's. This is a huge change from the distributed computing designs of the last 10 years.
    >
    > Imho, the question that will become increasingly important in the future - "Can a company
    > afford OS platforms for their future centralized, very HA strategy that have "one app,
    > one OS" App/ISV support cultures...


    Well, unless you're considering something like ERP or an office
    productivity suite a single app, then I guess I'd disagree.

    Server consolidation is happening and will continue to happen, but if
    OpenVMS doesn't run the applications that are being consolidated, why
    would OpenVMS even be a consideration?

    The "one app, one OS" scene hasn't played in years. Nobody I can think
    of today does that other than on existing older equipment that *will*
    be consolidated as equipment is replaced (or for very specialized
    applications on special purpose servers that *should* be kept
    isolated.)

    > and where the OS vendors release 5-20 security patches
    > per month?"
    >


    You really should stop using that argument, too. It's been rebutted in
    many previous threads.

    Windows and *nix are so much better today vs. the past, in every
    aspect, that anyone paying even a little attention can predict that
    they will only get better.

    > My advice to Cust's - think like Wayne Gretzky (ok, he's a hockey player) and the way
    > He became a great player .. "Do not skate to where the puck is, but where it will be in
    > the next play sequence .."
    >
    > :-)


    The puck looks to me like it's still heading towards x86. The article
    you reference supports this argument better than most I've read.
    Thanks.

    ----------
    And a semi-non-related news item just in:
    +
    "Nigeria overtakes Ireland as Guinness' second-largest market."
    +
    If anyone needed proof that the world is getting weirder, there it
    is!



  12. RE: The Common System Interface: Intel's Future Interconnect

    > -----Original Message-----
    > From: Doug Phillips [mailto:dphill46@netscape.net]
    > Sent: August 30, 2007 5:10 PM
    > To: Info-VAX@Mvb.Saic.Com
    > Subject: Re: The Common System Interface: Intel's Future Interconnect


    [snip]

    > >
    > > Remember the relative importance to Cust's in terms of both what they

    > have and what they
    > > need In the future:
    > >
    > > 1. App = 50-60%
    > > 2. OS = 25-35%
    > > 3. Server HW - 10-15%

    >
    > I really didn't want to get into this again, but let's talk about
    > those numbers. You say "relative importance to Cust's" of the App is
    > 50-60% ?? I say relative importance of the App is 100% and the OS &
    > hardware (and I include workstations, printers and all other
    > peripherals and infrastructure) fall into the category of "necessary
    > evils." Your numbers more closely reflect the rough cost of acquiring
    > the ability to perform the application.
    >


    Lets not get silly here ok?

    Very few Cust's run an application by itself. There are OS customizations like batch
    jobs, OS security on files, custom reports, DCL scripts (and their equivalent in UNIX
    world), ISV customizations in terms of support utilities, pre and post processing jobs,
    OS specific backups, data archiving, security auditing etc.

    These are all typically heavily integrated with the primary App's. on that system.

    When looking at new OS platforms, Cust's understand that these are all considerations
    that need to be taken into consideration.

    Saying the application priority is 100% is a tad naive.

    > Platform security, support, reliability, performance and all other OS/
    > hardware factors effect cost and are certainly important but they are
    > subordinate to and dependent upon the needs of the application.
    >
    > You don't buy an application you can't run, and you don't buy a
    > platform that won't run your application. Before anyone else says it,
    > I'll say it first: "Duh!"
    >


    Never said anything otherwise .. are you trying to make a point in there somewhere?

    > >
    > > While #3 gets all sorts of attention in techie newsgroups, in the big

    > picture,
    > > #1 and #2 are much more important to Cust's. With a massive glut in

    > available compute
    > > cycles in most Cust's environment today, Cust's are not impressed

    > with fantastic new
    > > computer speeds that will increase their glut of available compute

    > cycles even more.
    > >
    > > Btw, this applies to all platforms.
    > >
    > > Also, keep in mind that there is now a massive trend to consolidating

    > both servers and
    > > DC's. This is a huge change from the distributed computing designs of

    > the last 10 years.
    > >
    > > Imho, the question that will become increasingly important in the

    > future - "Can a company
    > > afford OS platforms for their future centralized, very HA strategy

    > that have "one app,
    > > one OS" App/ISV support cultures...

    >
    > Well, unless you're considering something like ERP or an office
    > productivity suite a single app, then I guess I'd disagree.
    >
    > Server consolidation is happening and will continue to happen, but if
    > OpenVMS doesn't run the applications that are being consolidated, why
    > would OpenVMS even be a consideration?
    >


    Consolidation within the same platform is what I was referring to. Unless it is a very
    simple IT service or application, changing OS Platforms during any server consolidation
    initiative raises the risk level from 2-3 to something like 7-8.

    > The "one app, one OS" scene hasn't played in years. Nobody I can think
    > of today does that other than on existing older equipment that *will*
    > be consolidated as equipment is replaced (or for very specialized
    > applications on special purpose servers that *should* be kept
    > isolated.)
    >


    Again, you have obviously not been in real world Operations or IT environments
    for awhile.

    Why is VMware so hot? Lots of CPU cycles are available, so Why not take 5-10 Windows
    Biz applications and put them on a single Windows OS instance?

    Because there is a One App, One OS culture and each group does not want their App
    sharing the same OS resources as someone else's. Even if both Apps run on servers that
    in their peak times only reach 10-15% CPU utilization. That's reality.

    VMware shares are going through the roof right now and this is the absolute
    biggest reason why this is happening. [Course, VMware does not actually reduce the
    work to maintain all the various OS instances, but that's another discussion.]

    > > and where the OS vendors release 5-20 security patches
    > > per month?"
    > >

    >
    > You really should stop using that argument, too. It's been rebutted in
    > many previous threads.
    >
    > Windows and *nix are so much better today vs. the past, in every
    > aspect, that anyone paying even a little attention can predict that
    > they will only get better.
    >


    [Note that I referred to Linux and Windows, not UNIX]

    Please, again your statement reflects only that you likely have an application focus
    with little DC Operations experience. And likely one that has been bitten by all the
    Linux and Windows hype. And likely not someone who has an environment with hundreds of
    Wintel/Linux servers that need security patches every month.

    Here is the Linux RH web site for security patches:
    https://www.redhat.com/archives/enterprise-watch-list/

    Now click on "thread" for each month and go back as many months as you want. Each
    and every month, there are 5-20 security patches (sometimes many more).

    Why not create a graph from 2003-2007 to show how the number of security patches
    released each month has gone down? [hint - you won't like the trend]

    What's to rebut?

    Yes, not every security patch applies to all systems, but most shops have no idea
    what all the services their development and production systems are using, and each
    readme file on the security patch is often vague (for good reason) as to all of the
    associated details around each patch. And many of these one line patch links are bundled
    security patches as well.

    So, saying the Windows (and Linux) server issues are getting any better is really
    laughable and is really only an indication of someone bitten by the hype bug.

    Microsoft still also releases numerous security patches each and every month. Here is
    a recent example:

    http://www.eweek.com/article2/0,1895,2127475,00.asp (May 2007)
    Microsoft has released patches for 19 vulnerabilities, 14 of which are critical, hitting
    at holes in Excel, Word, Office, Exchange, Internet Explorer, cryptographictechnology
    and the whopper of them all, the zero-day vulnerability in the DNS Server'suse of RPC.

    [and yes, you do need explorer on your Windows servers]

    http://www.theinquirer.net/?article=41698 (August 2007)
    SOFTWARE GIANT Microsoft has issued patches for nine security flaws, including six
    critical ones.

    ... "Four of the patches keep hackers from breaking into computers through Web pages and
    are rated critical. [snip] The other three updates are rated important and one of them
    is for the "super secure" Vista OS."

    > > My advice to Cust's - think like Wayne Gretzky (ok, he's a hockey

    > player) and the way
    > > He became a great player .. "Do not skate to where the puck is, but

    > where it will be in
    > > the next play sequence .."
    > >
    > > :-)

    >
    > The puck looks to me like it's still heading towards x86. The article
    > you reference supports this argument better than most I've read.
    > Thanks.
    >


    Once the hype bug has bitten someone, it is tough to get them back to reality.

    Hey, I know what it is like. Back in mid 90's, in addition to being an OpenVMS Ambassador,
    I used to also be part of the Digital Windows NT Wizards Program. At that time, there were
    many monthly Windows security patches, but even back then the std response was "hey, its
    getting better and soon Windows will take over the world.." (ok, even then I did not believe
    all of the hype, but the hype bug is tough to shake.)

    Regards


    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.






  13. Re: The Common System Interface: Intel's Future Interconnect

    Main, Kerry wrote:
    >>-----Original Message-----
    >>From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca]
    >>Sent: August 29, 2007 9:35 PM
    >>To: Info-VAX@Mvb.Saic.Com
    >>Subject: Re: The Common System Interface: Intel's Future Interconnect
    >>
    >>Main, Kerry wrote:
    >>
    >>>All,
    >>>
    >>>The following article may be of interest: (August 28, 2007)
    >>>http://www.realworldtech.com/page.cf...2807020032&p=1

    >>
    >>
    >>Mr Main,
    >>
    >>What will differentiate a 64 bit 8086 plugged into a CSI interface from
    >>a 64 bit IA64 also plugged into a CSI interface ?
    >>
    >>If both have access to the same type of memory, cache etc, then won't
    >>the industry standard architecture that has competition from AMD end up
    >>being far superior than some proprietary IA64 thing that requires its
    >>own proprietary funky compilers due to ist EPIC nature ?

    >
    >
    > JF -
    >
    > Remember the relative importance to Cust's in terms of both what they have and what they
    > need In the future:
    >
    > 1. App = 50-60%
    > 2. OS = 25-35%
    > 3. Server HW - 10-15%
    >
    > While #3 gets all sorts of attention in techie newsgroups, in the pig picture,
    > #1 and #2 are much more important to Cust's. With a massive glut in available compute
    > cycles in most Cust's environment today, Cust's are not impressed with fantastic new
    > computer speeds that will increase their glut of available compute cycles even more.
    >
    > Btw, this applies to all platforms.
    >
    > Also, keep in mind that there is now a massive trend to consolidating both servers and
    > DC's. This is a huge change from the distributed computing designs of the last 10 years.
    >
    > Imho, the question that will become increasingly important in the future - "Can a company
    > afford OS platforms for their future centralized, very HA strategy that have "one app,
    > one OS" App/ISV support cultures and where the OS vendors release 5-20 security patches
    > per month?"
    >
    > My advice to Cust's - think like Wayne Gretzky (ok, he's a hockey player) and the way
    > He became a great player .. "Do not skate to where the puck is, but where it will be in
    > the next play sequence .."
    >
    > :-)


    I've long noticed this cycle. For example, when I first
    started, one of our biggest customers was a large NY bank
    that had been burned by its centralized data processing
    operation. All their eggs in one basket, and someone had
    dropped the basket. (I don't know the details.)

    At any rate, when I started out, they were in the midst of
    distributing their computer resources out to the end users,
    i.e. the branches and departments of the bank. We (and many
    other vendors) were using RSTS/E systems for this. (Other
    vendors were basing their applications on other platforms,
    DG, Prime, Perkin-Elmer, ...) Many of the changes DEC made
    to RSTS/E V6B were specifically for this customer. Quite soon,
    they networked all these little systems together, since it
    was much more efficient than producing dozens of tapes to
    hand off each day and shuttle around the city.

    Eventually, this "let 100 flowers bloom" strategy resulted
    in too much (perceived) inefficiency, too much duplication
    of effort, too many operational and security issues (e.g.
    making sure all the systems correctly enforced the bank's
    security policies, proper backups were being done, etc.),
    and improve network efficiency by putting everything in one
    room on one Ethernet rather than having dozens of 56Kb
    leased lines, so they (and other of our customers) instituted
    a policy of consolidation. Get everything into a proper data
    center, make sure the systems were run by professionals, make sure
    operations procedures were documented and followed, and make
    sure proper testing and change procedures were followed.
    A lot of this consolidation was done onto VAX/VMS systems in
    the early '80s.

    This of course re-instituted the original problems that
    caused the distribution in the first place, such as a
    large, high-inertia IT structure that took a long time
    to make application changes and was run by people who
    understood the IT stuff much better than they understood
    the business, so often the changes didn't really address
    the problems they were intended to solve, or were "one
    step forward, two steps back", or took much longer to
    implement or cost the "customer" (i.e. the company's
    department that was the user of the application), far
    more than seemed reasonable. ("You mean you want to
    charge my department $$$$$ just to produce a summary
    page at the end of the report and email it to me twice
    a day?!?!?! And it's going to take 6 months to implement?
    Get real!")

    So when it became practical to move a big chunk of the
    data center into a small box in the corner, under the
    direct control of the user (i.e. a PC) and bypass the
    IT bureaucracy, that's exactly what happened.

    Now all the PC's are getting reconsolidated, which
    will eventually result in another redistribution,
    this time most likely to a (small) rack of blade
    servers in the corner with a (small) SAN array for
    storage. "Small" in this context means a few
    tens of multi-GHz processors and a few Terabytes
    of storage, i.e. orders of magnitude more powerful
    than the PCs which preceded them last time around,
    which were in turn orders of magnitude more powerful
    than the PDP-11/70's in the first cycle.

    The cycle keeps repeating. I expect it will go on
    for ever :-)


    > Regards
    >
    >
    > Kerry Main
    > Senior Consultant
    > HP Services Canada
    > Voice: 613-592-4660
    > Fax: 613-591-4477
    > kerryDOTmainAThpDOTcom
    > (remove the DOT's and AT)
    >
    > OpenVMS - the secure, multi-site OS that just works.
    >
    >
    >
    >
    >



    --
    John Santos
    Evans Griffiths & Hart, Inc.
    781-861-0670 ext 539

  14. Re: The Common System Interface: Intel's Future Interconnect

    Main, Kerry wrote:
    > Microsoft still also releases numerous security patches each and every month. Here is
    > a recent example:



    Until HP decides to release that MONITOR CLUSTER patch, you are in no
    position to talk. I'd much rather have patches come out rapidly than
    having to live with dysfunctional software that doesn't match what the
    SPD promised.

    > and the whopper of them all, the zero-day vulnerability in the DNS Server's use of RPC.



    Can't read beyond "use". But again, DNS isn't exactly your forte, since
    the owner of VMS hasn't given any indication of whether its BIND
    implementation has the same flaws as BIND on other platforms, flaws
    which have been made public and for which patches exist.


    Perhpas you might wish to compare VMS with MAC OS classic. (say 8.6).
    Not many new patches have been issued for 8.6 in recent years. Must be a
    stable, full features OS without security vulnerabilities.

    Or perhaps the idea behind lack of patches is the fact that MAC OS
    classic is "mature" and no longer developped and no new patches exist
    and very few modern applications still run on it ? Reminds me of VMS.

  15. RE: The Common System Interface: Intel's Future Interconnect

    > -----Original Message-----
    > From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca]
    > Sent: August 30, 2007 3:13 PM
    > To: Info-VAX@Mvb.Saic.Com
    > Subject: Re: The Common System Interface: Intel's Future Interconnect
    >
    > Main, Kerry wrote:
    > > You did not read my response as you viewed this as simply another

    > opportunity to flog
    > > your personal views that the primarily "One App, One OS" x86 server

    > HW platform is going
    > > to take over the server world like it did the desktop.

    >
    >
    > 8086 is not a "one app one server platform". Windows is. You can have a
    > serious 64 bit 8086 server running a serious operating system which
    > supports multiple applications.
    >


    Great theory ..what is the reality?

    What is the proportion of Windows/Linux servers on x86 servers vs. other OS
    Platforms on X86?

    Perhaps something like 85-95% range?

    >
    > The point is that the 8086 is coming of age and with common interfaces
    > with IA64, it will become possible to build large systems that have
    > similar or better performance with an 8086 compared to IA64. At that
    > point, there is no point in continuing to try to shove an unpopular
    > platform down customer's throats.
    >


    Again, you are missing the point. Great hardware goes no where without the OS
    culture and applications that go with it.

    How do you propose to solve the "One App, One OS" culture associated with Windows
    and Linux?

    Most of the Customers X86 servers today are running at 10-15% at peak times, so they
    do not need faster systems to drop this utilization to 5%. In order to reduce their
    FTE staffing counts that make up 60-70% of their IT budget, they need to consolidate
    their applications on fewer servers and fewer OS instances.

    > There is no reason HP couldn't build a CSI based 8086 with EFI console
    > with all the same bells and whistles as IA64 systems. And once this
    > happens, there is also no reason to continue to build IA64 based
    > systems.


    Been discussed many times by many people. See previous threads.

    Bottom line is that if a Cust wants Appx and Appx is available on OpenVMS, or is
    developing /porting AppY on OpenVMS because of the added security and HA value it brings,
    then whether the server is Integrity or x86 CPU's is a very small part of the equation.

    Regards


    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.




  16. RE: The Common System Interface: Intel's Future Interconnect

    > -----Original Message-----
    > From: JF Mezei [mailto:jfmezei.spamnot@vaxination.ca]
    > Sent: August 30, 2007 9:05 PM
    > To: Info-VAX@Mvb.Saic.Com
    > Subject: Re: The Common System Interface: Intel's Future Interconnect
    >
    > Main, Kerry wrote:
    > > Microsoft still also releases numerous security patches each and

    > every month. Here is
    > > a recent example:

    >
    >
    > Until HP decides to release that MONITOR CLUSTER patch, you are in no
    > position to talk. I'd much rather have patches come out rapidly than
    > having to live with dysfunctional software that doesn't match what the
    > SPD promised.
    >


    We are talking about *security* patches here. What security problem does the Monitor
    problem cause?

    > > and the whopper of them all, the zero-day vulnerability in the DNS

    > Server's use of RPC.
    >
    >
    > Can't read beyond "use". But again, DNS isn't exactly your forte, since
    > the owner of VMS hasn't given any indication of whether its BIND
    > implementation has the same flaws as BIND on other platforms, flaws
    > which have been made public and for which patches exist.
    >


    Same answer as before - as security issues are discovered (internally or externally),
    they are reviewed to determine if it really is a problem and if so, the potential
    impact, what the fix requires and how to distribute the patch.

    OpenVMS has an excellent reputation in this regard.

    >
    > Perhpas you might wish to compare VMS with MAC OS classic. (say 8.6).
    > Not many new patches have been issued for 8.6 in recent years. Must be
    > a
    > stable, full features OS without security vulnerabilities.
    >
    > Or perhaps the idea behind lack of patches is the fact that MAC OS
    > classic is "mature" and no longer developped and no new patches exist
    > and very few modern applications still run on it ? Reminds me of VMS.


    Don't know MAC's or even follow them.

    However, what little I have heard of them is that they seem to have pretty good
    security.

    Regards


    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.





  17. RE: The Common System Interface: Intel's Future Interconnect

    > -----Original Message-----
    > From: John Santos [mailto:john@egh.com]
    > Sent: August 30, 2007 8:59 PM
    > To: Info-VAX@Mvb.Saic.Com
    > Subject: Re: The Common System Interface: Intel's Future Interconnect
    >
    > Main, Kerry wrote:
    > >>-----Original Message-----


    [snip]

    > >
    > > Also, keep in mind that there is now a massive trend to consolidating

    > both servers and
    > > DC's. This is a huge change from the distributed computing designs of

    > the last 10 years.
    > >
    > > Imho, the question that will become increasingly important in the

    > future - "Can a company
    > > afford OS platforms for their future centralized, very HA strategy

    > that have "one app,
    > > one OS" App/ISV support cultures and where the OS vendors release 5-

    > 20 security patches
    > > per month?"
    > >
    > > My advice to Cust's - think like Wayne Gretzky (ok, he's a hockey

    > player) and the way
    > > He became a great player .. "Do not skate to where the puck is, but

    > where it will be in
    > > the next play sequence .."
    > >
    > > :-)

    >
    > I've long noticed this cycle. For example, when I first
    > started, one of our biggest customers was a large NY bank
    > that had been burned by its centralized data processing
    > operation. All their eggs in one basket, and someone had
    > dropped the basket. (I don't know the details.)
    >


    [snip]

    Re: consolidation - distributed computing cycle.

    Imho, what is now happening is a struggle to find the "happy medium". Neither of these
    strategies on their own is the right strategy in all cases.

    What seems to be happening more and more is centralized technology where itmakes sense
    (latency and bandwidth are restrictions that might require some servers left distributed),
    but dedicating IT Resources to deal with the Business Units i.e. IT resources who attend
    regular BU meetings and who's role is to assist the BU's understand how IT can help their
    bottom line by assisting them in providing more competitive and cost efficient solutions.

    This is designed to fix the "glass house does not understand our requirements" complaints
    Associated with previous centralized computing strategies.

    And make no mistake - IT groups today are motivated to ensure the BU's are happy, because
    they know they will be outsourced if they continue to ignore the BU's.

    Regards

    Kerry Main
    Senior Consultant
    HP Services Canada
    Voice: 613-592-4660
    Fax: 613-591-4477
    kerryDOTmainAThpDOTcom
    (remove the DOT's and AT)

    OpenVMS - the secure, multi-site OS that just works.







  18. Re: The Common System Interface: Intel's Future Interconnect

    Main, Kerry wrote:
    > What is the proportion of Windows/Linux servers on x86 servers vs. other OS
    > Platforms on X86?


    > How do you propose to solve the "One App, One OS" culture associated with Windows
    > and Linux?



    Like it or not, 8086 is the industry standard, and its future is NOT in
    question. IA64's future is in question, and with CSI, its future will
    continue to be in question, especially since the 8086 will scale even
    higher into IA64's market niche.

    VMS is hindered by running only on that IA64 thing. IA64 is not an asset
    it is a liability to VMS.


    If you want to attract new customers to VMS, you need VMS on a popular
    platform, one which will not generate question about its future. Porting
    VMS to a platform that everyone uses and which will exist in the long
    term would show HP's true commmitment to VMS, and would show the world
    HP intends to allow VMS to grow. Refusing to port VMS to 8086 sends a
    clear message that HP does not intend to continue VMS beyond IA64 (and
    when combined with Livermore's recent statements as well as HP's
    invitation for Cerner to abandon VMS in favour of HP-UX, VMS may spiral
    down faster and go down before IA64.


    CSI will give the 8086 capabilities once reserved to high end chips. And
    porting VMS to it would give VMS access to a far greater market.

    Of course, HP's actions are in fact resulting in a faster decline of
    VMS. If they had been interested in just maintaining the installed base,
    they would have continued Alpha sales. Forcing customers to go thorugh
    the hassles of porting to IA64 will accelerate the loss of VMS customers.

  19. Re: The Common System Interface: Intel's Future Interconnect

    If you ask IT people, they will know that VMS is a good quality
    operating system, that it has good documentation and great clustering
    capabilities.

    BUT: They will say they can't get VMS because it doesn't run their apps,
    it doesn't run on viable industry standard platform, and because there
    is no commitiment to VMS from the vendor.

    Porting VMS to the 8086 will solve the server image issue as well as the
    commitment issue. It won't magically bring all apps to VMS. But without
    this port, it ensures that VMS will not get new apps and continue to
    reduce the number of apps it has left.

    As long as HP makes public statements about just wanting to be able to
    convert the VMS installed based to HP-UX customers, it will send a
    strong signal that HP has no interest in growing VMS and as a result,
    ISVs will stay away from it and in doing so, will help HP downsize VMS
    out of existance.

    And whenever some HP employee supports their employer's bid to downsize
    VMS out of existance, it makes those customers loyal to VMS feel even
    more alone because they don't even see VMS staff trying to fight for
    VMS's success/survival within HP.

  20. Re: The Common System Interface: Intel's Future Interconnect

    On 08/30/07 22:44, JF Mezei wrote:
    [snip]
    >
    >
    > CSI will give the 8086 capabilities once reserved to high end chips. And
    > porting VMS to it would give VMS access to a far greater market.


    QuickPath (new name of Common System Interface) offers *nothing*
    that Athlon64 and Opteron using HyperTransport haven't had for 4 years.

    --
    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!

+ Reply to Thread
Page 1 of 2 1 2 LastLast