By: anonymous (anonymous.delete@this.email.cz), May 26, 2007 3:49 pm
Room: Moderated Discussions
Just my observation :
I am part of IT support for diverse universe of customers - Windows only machines , and from the considerable amount of machines that we are supporting less then 2% are 4 socket machines - and less then 0.5% of machines are 4x4(socket/cpu) servers - all Xeon MP Tulsa running some exotic business application. We aren't supporting even one ! 8 socket Windows based server.
OK we have some 4-5 4 socket machines running Xeon MP, and i faintly seem to remember few Pentium III 4 socket "rockets" but none of them are active anymore.
I encountered a truly 8 socket machine only two times - one was Opteron ESX Development server used by "insane Unix Guys" to test Windows clustering inside ESX - and boy did they have fun (me too) - it was truly perfect platform for simulation of how not to do the clustering ..., and the second one was P4 based, IBM X3 based i think, running some heavy DB... with everything that you can opt on that one - including some serious amount of memory.
As side note for the number of supported servers - its few thousands ... :)
And from i can see from my position - i personally don't see the 8 socket machines as the next big thing - at least not under Windows. All those precious 4 socket machines are running either DB which needs loads of memory, or some business application that i never heard before :) that seems to be able to load that particular server.
All around what i can see that people are migrating to blades - mainly 2 socket ones. 2 socket blade can run a lot these days, if its not enough or there is higher reliability requested - clustering of 2 socket systems. 4 socket one -> only if you need DB with a really lot of memory or you have application that can load all 4 sockets. Plus if it is 4 socket machine - all that i have seen come with a lot of bells - top bin cpu-s, large amounts of memory, RAID-s, Fibrechannels, multiple teamed NIC cards .... i don't remember seeing a base one configuration anywhere.
Simply nobody uses the base config - when its 4 socket its almost 99% time a rather serious machine which brings the Total Cost of machine into serious $$$.
So maybe the cost of platform will come down, the performance will go up, but under Windows for normal customer there is almost nothing that can run there with any significance - and as such machines come fully loaded the price will stay still insanely high.
So in niche markets the penetrations may go up but in mainstream i have my doubts whether there will by any penetration at all :)
Michael S (already5chosen@yahoo.com) on 5/23/07 wrote:
---------------------------
>Jose (no@teh.win) on 5/23/07 wrote:
>---------------------------
>>Michael S (already5chosen@yahoo.com) on 5/17/07 wrote:
>>---------------------------
>>>
>>>Are you theorizing or do you really have first hand experience with these wonderful commodity 4-socket servers?
>>>I personally never even see one, but I'm not in the IT.
>>Start here: http://www.supermicro.com/Aplus/system/Tower/
>>
>
>That's exactly what I am asking about. Are them any good?
>
>>Or, heck, are HP http://web.amd.com/en-us/partners/hp/server.aspx
>>
>
>Sorry, HP DL585 is not commodity. It is a product of non-trivial in house engineering.
>
>and Sun http://web.amd.com/en-us/partners/sun/server.aspx
>
>Sun Fire V40z is commodity. I don't remember Sun having big success selling those. X4600 is not commodity.
>
>>The trouble with "commodity" 8 socket servers is simply the sheer real estate involved.
>>You can't fit eight sockets on a conventional motherboard, so you're looking at
>>daughter cards... and already you've walked away from "commodity." Now, there are
>>ways it *could* go (HTX slots so standardized single- or dual-socket daughter cards
>>become a commodity of sorts) but the industry likes to go its own way until the
>>vendors finally realize (again) the benefits of standardization. And the 8 socket
>>market will be small (but lucrative) enough that that may never happen.
>
>You are sort of making my point.
>I'd say the problem exists even with 4-way servers when you try to build then "commodity-way"
>i.e. with everything on a single motherboard.
>
>BTW, it is interesting that Supermicro machines are towers. IMO, 4-way "commodity"
>towers are more likely to be good then their 3U or 4U counterparts. However, towers
>are not fashionable in big IT organizations and small IT organizations rarely need 4-ways to start with.
>
I am part of IT support for diverse universe of customers - Windows only machines , and from the considerable amount of machines that we are supporting less then 2% are 4 socket machines - and less then 0.5% of machines are 4x4(socket/cpu) servers - all Xeon MP Tulsa running some exotic business application. We aren't supporting even one ! 8 socket Windows based server.
OK we have some 4-5 4 socket machines running Xeon MP, and i faintly seem to remember few Pentium III 4 socket "rockets" but none of them are active anymore.
I encountered a truly 8 socket machine only two times - one was Opteron ESX Development server used by "insane Unix Guys" to test Windows clustering inside ESX - and boy did they have fun (me too) - it was truly perfect platform for simulation of how not to do the clustering ..., and the second one was P4 based, IBM X3 based i think, running some heavy DB... with everything that you can opt on that one - including some serious amount of memory.
As side note for the number of supported servers - its few thousands ... :)
And from i can see from my position - i personally don't see the 8 socket machines as the next big thing - at least not under Windows. All those precious 4 socket machines are running either DB which needs loads of memory, or some business application that i never heard before :) that seems to be able to load that particular server.
All around what i can see that people are migrating to blades - mainly 2 socket ones. 2 socket blade can run a lot these days, if its not enough or there is higher reliability requested - clustering of 2 socket systems. 4 socket one -> only if you need DB with a really lot of memory or you have application that can load all 4 sockets. Plus if it is 4 socket machine - all that i have seen come with a lot of bells - top bin cpu-s, large amounts of memory, RAID-s, Fibrechannels, multiple teamed NIC cards .... i don't remember seeing a base one configuration anywhere.
Simply nobody uses the base config - when its 4 socket its almost 99% time a rather serious machine which brings the Total Cost of machine into serious $$$.
So maybe the cost of platform will come down, the performance will go up, but under Windows for normal customer there is almost nothing that can run there with any significance - and as such machines come fully loaded the price will stay still insanely high.
So in niche markets the penetrations may go up but in mainstream i have my doubts whether there will by any penetration at all :)
Michael S (already5chosen@yahoo.com) on 5/23/07 wrote:
---------------------------
>Jose (no@teh.win) on 5/23/07 wrote:
>---------------------------
>>Michael S (already5chosen@yahoo.com) on 5/17/07 wrote:
>>---------------------------
>>>
>>>Are you theorizing or do you really have first hand experience with these wonderful commodity 4-socket servers?
>>>I personally never even see one, but I'm not in the IT.
>>Start here: http://www.supermicro.com/Aplus/system/Tower/
>>
>
>That's exactly what I am asking about. Are them any good?
>
>>Or, heck, are HP http://web.amd.com/en-us/partners/hp/server.aspx
>>
>
>Sorry, HP DL585 is not commodity. It is a product of non-trivial in house engineering.
>
>and Sun http://web.amd.com/en-us/partners/sun/server.aspx
>
>Sun Fire V40z is commodity. I don't remember Sun having big success selling those. X4600 is not commodity.
>
>>The trouble with "commodity" 8 socket servers is simply the sheer real estate involved.
>>You can't fit eight sockets on a conventional motherboard, so you're looking at
>>daughter cards... and already you've walked away from "commodity." Now, there are
>>ways it *could* go (HTX slots so standardized single- or dual-socket daughter cards
>>become a commodity of sorts) but the industry likes to go its own way until the
>>vendors finally realize (again) the benefits of standardization. And the 8 socket
>>market will be small (but lucrative) enough that that may never happen.
>
>You are sort of making my point.
>I'd say the problem exists even with 4-way servers when you try to build then "commodity-way"
>i.e. with everything on a single motherboard.
>
>BTW, it is interesting that Supermicro machines are towers. IMO, 4-way "commodity"
>towers are more likely to be good then their 3U or 4U counterparts. However, towers
>are not fashionable in big IT organizations and small IT organizations rarely need 4-ways to start with.
>
Topic | Posted By | Date |
---|---|---|
Barcelona Article Online | David Kanter | 2007/05/16 03:20 AM |
Barcelona Article Online | PiedPiper | 2007/05/16 05:12 AM |
Yes, I left out a sentence there. Fixed (NT) | David Kanter | 2007/05/16 12:07 PM |
Barcelona Article Online | anonymous | 2007/05/16 06:01 AM |
Barcelona Article Online | Anonymous | 2007/05/16 06:28 PM |
Barcelona Article Online | anonymous | 2007/05/16 07:52 PM |
Barcelona Article Online | Anonymous1 | 2007/05/16 07:08 AM |
Barcelona Article Online | Dean M | 2007/05/16 11:09 AM |
Barcelona Article Online | David Kanter | 2007/05/16 12:38 PM |
Barcelona Article Online | Dean M | 2007/05/16 02:10 PM |
Barcelona Article Online | IntelUser2000 | 2007/05/16 02:59 PM |
Barcelona Article Online | Linus Torvalds | 2007/05/16 03:24 PM |
Barcelona Article Online | David Kanter | 2007/05/16 04:57 PM |
Barcelona Article Online | Michael S | 2007/05/17 05:07 AM |
Barcelona Article Online | IntelUser2000 | 2007/05/18 08:58 PM |
8 socket servers | Doug Siebert | 2007/05/16 04:58 PM |
8 socket servers | Michael S | 2007/05/17 05:20 AM |
8 socket servers | Joe Chang | 2007/05/17 07:38 AM |
8 socket servers | Alex Jones | 2007/05/17 09:35 AM |
8 socket servers | Jose | 2007/05/23 08:23 AM |
8 socket servers | Michael S | 2007/05/23 11:37 AM |
8 socket servers | anonymous | 2007/05/26 03:49 PM |
8 socket servers | Joe Chang | 2007/05/27 01:46 PM |
8 socket servers | Doug Siebert | 2007/05/23 09:56 PM |
8 socket servers | Joe Chang | 2007/05/24 04:33 AM |
8 socket servers | Anonymous | 2007/05/24 11:18 AM |
8 socket servers | Doug Siebert | 2007/05/24 10:47 PM |
8 socket servers | Linus Torvalds | 2007/05/25 10:35 AM |
8 socket servers | Nick | 2007/05/25 02:29 AM |
Performance estimation seems odd | Hotar | 2007/05/17 01:54 AM |
Performance estimation seems odd | David Kanter | 2007/05/17 08:38 AM |
microops vs macroops on page 4 | Peter Lund | 2007/05/17 12:04 PM |
microops vs macroops on page 4 | David Kanter | 2007/05/21 04:51 PM |
microops vs macroops on page 4 | EduardoS | 2007/05/21 05:42 PM |
microops vs macroops on page 4 | dess | 2007/05/21 07:00 PM |
Barcelona Article Online | Peter Lund | 2007/05/17 12:25 PM |
macro-op vs. micro-op | dess | 2007/05/21 07:24 AM |
macro-op vs. micro-op | David Kanter | 2007/05/21 04:38 PM |
macro-op vs. micro-op | dess | 2007/05/21 06:15 PM |
macro-op vs. micro-op | David Kanter | 2007/05/22 12:11 AM |
macro-op vs. micro-op | dess | 2007/05/22 03:56 AM |
macro-op vs. micro-op | Gipsel | 2007/05/22 05:05 AM |
macro-op vs. micro-op | dess | 2007/05/22 05:52 AM |
macro-op vs. micro-op | anonymous | 2007/05/22 06:14 AM |
macro-op vs. micro-op | dess | 2007/05/22 06:44 AM |
macro-op vs. micro-op | EduardoS | 2007/05/22 02:19 PM |
macro-op vs. micro-op | dess | 2007/05/24 08:52 AM |
Stop comparing apples to oranges | EduardoS | 2007/05/22 02:30 PM |
Stop comparing apples to oranges | dess | 2007/05/22 04:09 PM |
Stop comparing apples to oranges | dess | 2007/05/22 04:30 PM |
Stop comparing apples to oranges | EduardoS | 2007/05/22 04:31 PM |
Stop comparing... apples to oranges? | dess | 2007/05/24 09:30 AM |
Stop comparing apples to oranges | anonymous | 2007/05/22 08:12 PM |
Stop comparing apples to oranges | EduardoS | 2007/05/23 02:50 PM |
macro-op vs. micro-op | anonymous | 2007/05/22 06:08 AM |
macro-op vs. micro-op | dess | 2007/05/22 06:40 AM |
macro-op vs. micro-op | anonymous | 2007/05/22 06:48 AM |
macro-op vs. micro-op | dess | 2007/05/21 08:30 PM |
macro-op vs. micro-op | anonymous | 2007/05/22 06:44 AM |
macro-op vs. micro-op | dess | 2007/05/24 09:38 AM |
macro-op vs. micro-op | Michael S | 2007/05/22 05:26 AM |