Comments for LooseBolts http://loosebolts.wordpress.com Thoughts on Cloud Computing, Infrastructure at Scale, and Data Centers Thu, 19 Nov 2009 19:04:44 +0000 http://wordpress.com/ hourly 1 Comment on A Practical Guide to the Early Days of Data Center Containers by A Standard for Data Center Containers? | Nexus Technologies - IT Company providing Virtualisation Solutions, IT managed Services and IT Consultancy http://loosebolts.wordpress.com/2009/10/29/a-practical-guide-to-the-early-days-of-data-center-containers/#comment-556 A Standard for Data Center Containers? | Nexus Technologies - IT Company providing Virtualisation Solutions, IT managed Services and IT Consultancy Thu, 19 Nov 2009 19:04:44 +0000 http://loosebolts.wordpress.com/2009/10/29/a-practical-guide-to-the-early-days-of-data-center-containers/#comment-556 [...] is no set industry standards when it comes to data center containers,” Manos wrote in a recent blog post. “This means that each vendor might have their own approach on what goes in, and what stays out [...] [...] is no set industry standards when it comes to data center containers,” Manos wrote in a recent blog post. “This means that each vendor might have their own approach on what goes in, and what stays out [...]

]]> Comment on Support UnderArmour / Support our Veterans by Nigel http://loosebolts.wordpress.com/2009/11/19/support-underarmour-support-our-veterans/#comment-555 Nigel Thu, 19 Nov 2009 12:48:58 +0000 http://loosebolts.wordpress.com/2009/11/19/support-underarmour-support-our-veterans/#comment-555 Michael, I was quite surprised to see you wearing a poppy on stage for your talks in London. I struck me interesting that here is an American wearing a Poppy on Rememberance Day. Especially in comparison to many of our own native speakers who went poppy-less. Nigel Michael,
I was quite surprised to see you wearing a poppy on stage for your talks in London. I struck me interesting that here is an American wearing a Poppy on Rememberance Day. Especially in comparison to many of our own native speakers who went poppy-less.

Nigel

]]> Comment on Data Center Junk Science: Thermal Shock \ Cooling Shock by Chris http://loosebolts.wordpress.com/2009/10/01/data-center-junk-science-thermal-shock-cooling-shock/#comment-554 Chris Tue, 17 Nov 2009 11:28:28 +0000 http://loosebolts.wordpress.com/2009/10/01/data-center-junk-science-thermal-shock-cooling-shock/#comment-554 "Since the group that wrote this book included all of the major server vendors, it must have been created with some type of justifiable reason." My guess is that it's the same reason that air temps must be at 17C. Consensus and covering ones own ass. If you were being scientific about it you'd strap a dozen thermocouples to say 10000-20000 servers and actually measure the temperature excursion vs failure rate in half a dozen facilities over the life of the servers. My guess is that unless Google or Mr Manos set this experiment up then it simply hasn't been done. “Since the group that wrote this book included all of the major server vendors, it must have been created with some type of justifiable reason.”

My guess is that it’s the same reason that air temps must be at 17C. Consensus and covering ones own ass.

If you were being scientific about it you’d strap a dozen thermocouples to say 10000-20000 servers and actually measure the temperature excursion vs failure rate in half a dozen facilities over the life of the servers. My guess is that unless Google or Mr Manos set this experiment up then it simply hasn’t been done.

]]> Comment on Data Center Junk Science: Thermal Shock \ Cooling Shock by Chris http://loosebolts.wordpress.com/2009/10/01/data-center-junk-science-thermal-shock-cooling-shock/#comment-553 Chris Tue, 17 Nov 2009 11:23:07 +0000 http://loosebolts.wordpress.com/2009/10/01/data-center-junk-science-thermal-shock-cooling-shock/#comment-553 "In data centres it is the temperature – where is it measured?" This might be a little cynical, but I'd guess it's at the instrument which shows the lowest excursion. “In data centres it is the temperature – where is it measured?”

This might be a little cynical, but I’d guess it’s at the instrument which shows the lowest excursion.

]]> Comment on Data Center Junk Science: Thermal Shock \ Cooling Shock by NEBSman http://loosebolts.wordpress.com/2009/10/01/data-center-junk-science-thermal-shock-cooling-shock/#comment-551 NEBSman Sat, 07 Nov 2009 20:38:20 +0000 http://loosebolts.wordpress.com/2009/10/01/data-center-junk-science-thermal-shock-cooling-shock/#comment-551 Mike, as well as Dave are right. 3C/hr is not a shock, it is insignificant change on temperature at glacier pace. Mil-810 shock is a shock, and can have some real reliability issues, but how many such events you expect to see in the data center ? NEBS specifications call for equipment to handle temperature rates of change of 30C/hr. As far as I know (and been on designing NEBS compliant equipment, including bu t not limited to servers for quite some time), we have newer had any real issues meeting this requirement on tests, or any failures on field attributed to same. Now, recall that the ASHRAE stands for "American Society of Heating, Refrigerating and Air Conditioning Engineers". Sure, there are server vendors in the bunch, and all are happy to overspecify the environmentals -- server vendors "just in case", and to maybe save a few $$s on cooling solution cost (applies to max temp's not changes). "R" people want to keep the box servers live within as small as possible -- bigger the box, less revenue and profit there is to be had. Do not expect objective environmental specifications come out of TC9.9. The green movement is good here -- it calls for change, or at least forces people to defend the state of affair by data, which "keep the box small" folks will not have to offer. EU CoC for data centers refers to yet to be published revision ETSI environmental spec for the IT equipment. This calls for Environmental category 3.1 (tops at 40C), which includes 3.1E (excursion to 45C) for limited time periods. Btw, the only salient revision to present spec is addition of two words "Data Centres" and "Computer Halls" under list of where the class 3.1 applies, right after "Telecommunications Centres" and "storage rooms for valuable and sensitive products". ETSI Spec is ETSI EN 300 019-1-3, and can be downloaded for free from ETSI web site. Mike, as well as Dave are right. 3C/hr is not a shock, it is insignificant change on temperature at glacier pace. Mil-810 shock is a shock, and can have some real reliability issues, but how many such events you expect to see in the data center ?

NEBS specifications call for equipment to handle temperature rates of change of 30C/hr. As far as I know (and been on designing NEBS compliant equipment, including bu t not limited to servers for quite some time), we have newer had any real issues meeting this requirement on tests, or any failures on field attributed to same.

Now, recall that the ASHRAE stands for “American Society of Heating, Refrigerating and Air Conditioning Engineers”. Sure, there are server vendors in the bunch, and all are happy to overspecify the environmentals — server vendors “just in case”, and to maybe save a few $$s on cooling solution cost (applies to max temp’s not changes). “R” people want to keep the box servers live within as small as possible — bigger the box, less revenue and profit there is to be had. Do not expect objective environmental specifications come out of TC9.9.

The green movement is good here — it calls for change, or at least forces people to defend the state of affair by data, which “keep the box small” folks will not have to offer.

EU CoC for data centers refers to yet to be published revision ETSI environmental spec for the IT equipment. This calls for Environmental category 3.1 (tops at 40C), which includes 3.1E (excursion to 45C) for limited time periods. Btw, the only salient revision to present spec is addition of two words “Data Centres” and “Computer Halls” under list of where the class 3.1 applies, right after “Telecommunications Centres” and “storage rooms for valuable and sensitive products”. ETSI Spec is ETSI EN 300 019-1-3, and can be downloaded for free from ETSI web site.

]]> Comment on Our Vision for Generation 4 Modular Data Centers – One way of Getting it just right . . . by dia a dia, bit a bit… por Silvio Meira » informaticidade: brasil investe em data centers http://loosebolts.wordpress.com/2008/12/02/our-vision-for-generation-4-modular-data-centers-one-way-of-getting-it-just-right/#comment-550 dia a dia, bit a bit… por Silvio Meira » informaticidade: brasil investe em data centers Thu, 05 Nov 2009 20:20:44 +0000 http://loosebolts.wordpress.com/2008/12/02/our-vision-for-generation-4-modular-data-centers-one-way-of-getting-it-just-right/#comment-550 [...] projeto, construção e operação de um centro computacional deste porte, vá ler este link, onde a microsoft explica a estratégia de centros computacionais da empresa. e onde [...] [...] projeto, construção e operação de um centro computacional deste porte, vá ler este link, onde a microsoft explica a estratégia de centros computacionais da empresa. e onde [...]

]]> Comment on Our Vision for Generation 4 Modular Data Centers – One way of Getting it just right . . . by dia a dia, bit a bit… por Silvio Meira » informaticidade: brasil investe em data centers http://loosebolts.wordpress.com/2008/12/02/our-vision-for-generation-4-modular-data-centers-one-way-of-getting-it-just-right/#comment-549 dia a dia, bit a bit… por Silvio Meira » informaticidade: brasil investe em data centers Thu, 05 Nov 2009 20:10:24 +0000 http://loosebolts.wordpress.com/2008/12/02/our-vision-for-generation-4-modular-data-centers-one-way-of-getting-it-just-right/#comment-549 [...] da complexidade do projeto, construção e operação de um centro computacional deste porte, vá ler este link, onde a microsoft explica a estratégia de centros computacionais da empresa. e onde há diagramas [...] [...] da complexidade do projeto, construção e operação de um centro computacional deste porte, vá ler este link, onde a microsoft explica a estratégia de centros computacionais da empresa. e onde há diagramas [...]

]]> Comment on Our Vision for Generation 4 Modular Data Centers – One way of Getting it just right . . . by dia a dia, bit a bit… por Silvio Meira » Brasil investe em data centers http://loosebolts.wordpress.com/2008/12/02/our-vision-for-generation-4-modular-data-centers-one-way-of-getting-it-just-right/#comment-548 dia a dia, bit a bit… por Silvio Meira » Brasil investe em data centers Thu, 05 Nov 2009 16:41:54 +0000 http://loosebolts.wordpress.com/2008/12/02/our-vision-for-generation-4-modular-data-centers-one-way-of-getting-it-just-right/#comment-548 [...] projeto, construção e operação de um centro computacional deste porte, vá ler este link, onde a microsoft explica a estratégia de centros computacionais da empresa. e onde [...] [...] projeto, construção e operação de um centro computacional deste porte, vá ler este link, onde a microsoft explica a estratégia de centros computacionais da empresa. e onde [...]

]]> Comment on The Cloud Politic – How Regulation, Taxes, and National Borders are shaping the infrastructure of the cloud by Your important business info going overseas? « BeYourOwnIT's Blog http://loosebolts.wordpress.com/2009/08/06/the-cloud-politic-how-regulation-taxes-and-national-borders-are-shaping-the-infrastructure-of-the-cloud/#comment-546 Your important business info going overseas? « BeYourOwnIT's Blog Mon, 02 Nov 2009 19:41:12 +0000 http://loosebolts.wordpress.com/2009/08/06/the-cloud-politic-how-regulation-taxes-and-national-borders-are-shaping-the-infrastructure-of-the-cloud/#comment-546 [...] fact that, once you put your sensitive personal and business data into the cloud, you really have no control over the physical location where that data resides. It’s very easy for large technology companies [...] [...] fact that, once you put your sensitive personal and business data into the cloud, you really have no control over the physical location where that data resides. It’s very easy for large technology companies [...]

]]> Comment on A Practical Guide to the Early Days of Data Center Containers by mmanos http://loosebolts.wordpress.com/2009/10/29/a-practical-guide-to-the-early-days-of-data-center-containers/#comment-545 mmanos Sat, 31 Oct 2009 02:50:03 +0000 http://loosebolts.wordpress.com/2009/10/29/a-practical-guide-to-the-early-days-of-data-center-containers/#comment-545 Greg, You are exactly right as far as the metal enclosure component of a data center container. The trick will be whether or not the change of use is significant enough for the IRS to either create a sub-class or another class entirely. Another option could be to keep the Container at 5 years and then individually depreciate the components inside. I guess my primary point here is that your Finance teams should definitely start looking at this as its bit trickier. As the container industry matures this should eventually be solved. Until then, make sure you have a bean-counter involved. (No offense of course :) ) \Mm Greg,

You are exactly right as far as the metal enclosure component of a data center container. The trick will be whether or not the change of use is significant enough for the IRS to either create a sub-class or another class entirely. Another option could be to keep the Container at 5 years and then individually depreciate the components inside. I guess my primary point here is that your Finance teams should definitely start looking at this as its bit trickier. As the container industry matures this should eventually be solved. Until then, make sure you have a bean-counter involved. (No offense of course :) )

\Mm

]]>