Interesting but.. The absence of an event does not mean the burden your policies place on the end users was necessary.
So you are saying you would not grant a waiver to a blanket full disk encryption policy for a lab that had higher performance needs and no sensitive data? Perhaps your policies are written better than the institute where the submitter works.
Blanket security policies with no procedure to obtain waivers are nearly always bad and are generally indicative of an IT organization that is poorly managed and not designed to meet the needs of the user community.
iPhone with unlimted data costs me less than BlackJack with unlimted data. The only difference between either of them and standard plan is the data fee.
You are misinformed.
There is nothing that would suggest a privacy problem at Haven, or that they would hand over your users information to any authorities. It seems you are concerned about things they do not wish to have on their servers. Since I assume you as a reasonable person do not actually see a "Privacy" problem in Haven. We must assume your objections are based on the types of content they do not allow (Which has nothing at all to do with privacy).
You will be hard pressed to find a host that will host pornography that is illegal in the EU (or the us) or actually hosting other peoples copyrighted material. (Not trackers or links by the way, their policy says the material itself).
I am not sure why anyone would want to help you. I am sure their are Slash dotters into kiddy porn, but it most be a very small number,
On the apple you just close the lid.. Not press the button, wait 5 seconds and close the lid. Most (I am sure there must be one that actually works somewhere) "PC" laptops will stay on if you just close the lid (even with sleep when lid is closed selected in windows) and then get really hot, overheat and shut down.
The patents don't lock you into amazon any more than you may have been without the patents. If you forsake all software that has ever applied for a US software patent, your life might get a little dull. Public companies must make all reasonable attempts to protect their intellectual property, it is part of their fiduciary responsibility to their shareholders. Not applying for a patent on something that was developed internally at a publicly traded company is just irresponsible.
You also have no idea what Amazon S3 is. Please do not treat slashdot comments you have read as facts, go to Amazon and read the S3 FAQ. Web Service != Web Interface.
Gdrive , thanks for the laugh.
It is not a web interface it is a Web Service. Webmin, NFS and AFS are all so entirely unrelated to any actual discussion of the S3 service it is not worth any further comment. RTFA then , RTFP, then go over to Amazon and RTFFAQ.
I have used sprint verizon EV-DO for over a year in about 25-30 US cities and canceled my service when I got my blackjack in Febrary and realized the tethered Blackjack on EDGE or 3G is quite a but faster than the EV-DO service. Theory is one thing and practical implementation is another entirely. At least in Dallas, Houston, San-Diego, Los Angeles, New York, Washington, Miami, Orlando, Chicago, Phoenix, Jackson, Denver and Philly the coverage, speed and reliability of the Sprint/Verizon network to be nearly as good as I receive on my phone with bluetooth PAN.
I realize a lot of those markets are 3g, but 3g is generally limited to the core of the market and even in outlying areas like my house that only has EDGE service, the Cingular network performs considerably better.
Slingplayer is unusable on EV-DO network and can work pretty well on the EDGE network. Perhaps if you are just doing something bursty like web traffic , you might be ok with EV-Do, might even think it is better , but for sustained bit rate I have never seen that to be the case.
I think most users will not see a huge difference between the network speeds. I have been using my blackjack since february. Tethered (bt pan) 3G generally tests out 280-350k range and EDGE in the 100-150k. Both have bursts and valleys (! have seen sling player on tethered pc jump over 600k for a bit on 3g. My house is edge only, it is fine for web browsing and even the mobile slingplayer. Many people seem to mistakenly think, cingulars edge network is the same as that terrible mess you get access to with a Verizon/sprint edge card. Tethered cingular is much faster and more reliable than anything ever experienced with one of those cards.
Battery life on 3g would be a huge problem for most consumers, the 3g drain is continuous when connected to a 3g network, not just when transfering data. 3G is just not ready for the mass market. I know we like to think everyone is like us, but the iphone was not made for the nerd community.
You are correct though.. A significant portion (likely somewhere between most and almost all) of the people on the Internet where someone is labeled as a fan boy are simply people trying to correct information that is factually incorrect. It does not mean the poster is mentally ill, a zealot, insecure or anything else. It simply means they were interested in injecting facts in a discussion on the Internet. Unfortunately for them, most people on the Internet are not interested in facts...
Which is why I provided you with real world examples you conveniently ignored, and chose to go with your own made up numbers that have no basis in reality. You just have no grasp at all of power consumption in a data center environment works and you refuse to look at the facts that are presented to you.
Please link us a real world example of a virtualization project where they experienced an actual increase of their electirc bill after the consolidation/virtualization was increased. I linked several that showed a significant decrease. You will not find any, because it does not happen. (Numbers you continue to just make up randomly in your head notwithstanding).
I can tell you 5 things about yourself:
1. You have never participated in a data center virtualization project of any size
2. You have never participated in a data center migration project of any size
3. You have never worked in a data center above the level of an untrained operator
4. You have never designed or evaluated any data center or significant server infrastructure.
5. You are incapable of admitting you are wrong.
Maybe if you can find the professional help you need to fix #5, you can learn how 1-4 actually work.
More reading you will ignore..I am sure if you actually read some of this , you would just stop posting..
Overall, consolidation with virtualization and blades can offer significant help with both space and power and cooling, but awareness of power/cooling requirements and careful planning are essential.
It's tempting to say that after long periods of testing, an increasing number of organizations are seeing the benefits of running applications on virtual servers housed in fewer physical boxes: increased resource utilization, faster server implementation, fewer devices to manage, lower management costs, a smaller data center footprint, and lower power and cooling costs.
Processor virtualization allows concurrent operating system execution environments to co-exist and share a fixed set of hardware resources. One of its many advantages is that this technology facilitates server consolidation, reducing both operating costs and power consumption.
"Data center power demands are growing at an unsustainable rate," said Bill Zeitler, senior vice president, IBM Systems and Technology Group. "The most important thing now is utilization of servers through technologies like virtualization."
I am going to respond because you are just a moron! Not total per rack TOTAL! Lets say you have 100 servers and they are in 10 RACKS. You consolidate these 100 Servers in 10 racks through virtualization. You now have 10 servers in 1 rack, carrying the load of what was 100 servers in 10 racks. The total power for that 1 rack will go up! The total power overall will go down, EVERY SINGLE TIME. You win the award for the densest person on the Internet. I have to assume you are just fucking with me at this point because no one who can type can be as idiotic as you. (I gave you an out , you might want to take it).
AC Costs absolutely become different! Some systems can not handle the higher density BTU load properly! But who cares because your assertion all along is that total power consumption goes up, not "Your current cooling system might need to be redesigned to accommodate the denser heat output". If you would like to discuss the impact of virtualization on Data Center cooling , I would be happy to. First you need to let us all know that you have finally grasped the fact that total power consumption does not go up. Until you can grasp that fact there is no point in trying to help you understand anything else.
The fact you think you need an additional cooling system proves you do not understand data center cooling either, 9/10 times the problem can be addressed with proper planning of the rack layout and its relationship to the existing CRAC units. Want to know why? Because Total BTUs went down right along with the Total Power. You of course do not believe using less power overall actually is a reduction in total power consumption. I am sure you will now link me some article you misread that you believe states virtualization increases the total BTUs in the datacenter.
You are just grasping at straws, trying to save some face. Man up, admit you were wrong and move on.
You like to read stuff on the Internet. You seem to focus on writers speaking hypothetically. Here are some real people talking about their real power and btu reductions:
You are showing your lack of first hand knowledge on the subject. Although many data centers still have 2 20-amp circuits per rack, you will find almost none built out in the last 5-10 years with less than 2-30amp circuits for racks and power distribution is generally built much more flexibly these days to support virtualization, blade servers and other high density configurations.
You need to stop getting all your information from Internet articles. They are generally written by writers and not experts in the Technology they are writing about. Anyone who built a new data center today only designed to handled 2-20amp circuits per rack would be fired.
Really not worth continuing trying to discuss this with someone who can not read. I understand it is difficult for someone with your very limited technical knowledge to make sense of all this.
The second article you linked also say "a physical server" nothing about increased overall power consumption.
One of the first problems often overlooked is the high power consumption and high heat output of a physical server hosting multiple virtualized servers. Remember that virtualized servers cause the hardware to run at a higher utilization rate, which means the power required to operate the physical server hardware goes up, as does the heat output.
A physical server running many virtual machines may run at a constant 80% or higher utilization rate, at which point the hardware is pulling quite a bit more power than a non-virtualized server running on average around 15% utilization.
The point is of course that many data centers are not designed to handle the increased power in the smaller footprint and may not be equipped to deal with the increased heat. It is all right there if your are able to read and comprehend. No where does it state overall power consumption will be increased. I am not going to discuss this with you any more because it is very clear you are simply incapable of admitting you have no idea at all what you are talking about. All you do is continue to link and misquote articles you did not understand to begin with.
The only appropriate response from you at this point is "I am sorry, I misread both of the articles I linked" anything else is just going to make you look like more of an ass than you already do. It is not a situation where you can agree to disagree or say it is a matter of opinion. You are just wrong.
You are clearly not able to understand what "a physical server" is. Older data centers may not be equipped to handle the increased power/btu per square foot of virtualization (or blades for that matter). This is the hidden cost your last article refers to, it has nothing at all to do with increased overall power consumption.
I can see where someone with no background in data center technology might fail to make the distinction, thats ok I am sure you are good at whatever it is you actually do.
If there is any more of this discussion you are talking to yourself, it is impossible to teach someone who believes they know everything.
You know all the author was trying to point out is that you should not base your projected savings on the 20W number, I am not sure why you think it meant something else.
Again you just misunderstand. The footprint reduction is what it is. Power consumption does not jump from 20W to 280W magically, it is directly related to the increase in utilization.
At this point it would be appropriate to interject, "all other things being equal"--meaning that there is no difference in application performance--a big assumption, and that the identical servers are used before and after
Type of servers used before and after = irrelevant. All things being equal (As I stated) is about application load not changing. because that has nothing to do with the virtualization process.
I am starting to feel sorry for you now.
I can tell you after virtualizing thousands of servers I have never seen a case where total power consumption has gone up. Working with IT personnel in dozens of large companies involved in virtualization projects, I have never heard even one suggest it might be happening to them.
You seem to believe a lot of things to be true, but none of them are based on anything you have actually done. I am sure this is the point where you claim I am full of crap. It does not matter if you believe me or not, because the facts speak for themselves and all you have to stand on is a really poorly written article by someone only slightly better informed than you.
Except 5 has nothing to do with what he said, you can't take my 5 and his 20 and 280 and think you have solved a riddle. You really are just embarrassing yourself now. What you just claimed would happen is not what happens in reality , period. There is no place in that article where the author tells you there will be an overall increase in power used. Period, end of discussion. Do you know why there is no place in that article where it says that? Because it would not happen. At no point does the author say total consumption before virtualization was 100W, you made that up based on a completely unrelated example I made.
One thing I can tell you based on the information provided in the article. Total power used before virtualization was more than 280W . Show me a direct quote from the article where it says total power used before virtualization was less than 280W. (You won't find it). I am really sorry I have proved to you multiple times that you misunderstood (and continue to misunderstand) the article.
You obviously know absolutely nothing about this subject, I would stop replying if watching you make silly statements was not so dang entertaining. Mark Twain would be happy to see you still proving his point so many years later.
Actually the fact they are not linear is exactly why you are wrong. The overhead in operating additional machines is exactly the difference. You should really stop arguing about something you clearly do not understand. It was obvious you did not understand the article or what he meant by footprint. To call me foolish for pointing out your own misunderstanding only serves to prove my point.
"In real power terms, a 300 watt server which was running at 20 watts is actually now running at 280 watts. You are reducing the footprint but putting in more processing power, so the power per footprint has gone up,"
To clarify what he is saying for you..He is saying the power per "FOOTPRINT" is going to go up, not the power per OS Instance. That is a no brainer, but he is being extremely vague in order to attempt to make a point. The power per FOOTPRINT is the power per Physical box. Wow! It is going up? Absolutely!
The number of physical boxes is going down. The author (Who apparently also has no experience actually implementing this) either doesn't get it, or was intentionally misleading to try and make a sensational point.
Reducing physical servers through virtualization will always reduce overall power consumption. You can continue to try to deny it all you want, but you really are just misunderstanding the article.
Somehow I doubt you will post here and tell us all that you did in fact misunderstand what he meant by footprint.
Slashdot Mirror
Interesting but.. The absence of an event does not mean the burden your policies place on the end users was necessary. So you are saying you would not grant a waiver to a blanket full disk encryption policy for a lab that had higher performance needs and no sensitive data? Perhaps your policies are written better than the institute where the submitter works. Blanket security policies with no procedure to obtain waivers are nearly always bad and are generally indicative of an IT organization that is poorly managed and not designed to meet the needs of the user community.
Did you read the article. The only unit tested that was cheaper was the Dell and it had vastly inferior graphics and a slower CPU...
Just because Dell can not build a clickable trackpad that works properly, does not mean it can not be done...
iPhone with unlimted data costs me less than BlackJack with unlimted data. The only difference between either of them and standard plan is the data fee. You are misinformed.
Yes it is definitely better to have no deisgn at all than have a design that will look dated 10 years after it's useful life....
Are you a child pornographer or a Warez l0zer?
There is nothing that would suggest a privacy problem at Haven, or that they would hand over your users information to any authorities. It seems you are concerned about things they do not wish to have on their servers. Since I assume you as a reasonable person do not actually see a "Privacy" problem in Haven. We must assume your objections are based on the types of content they do not allow (Which has nothing at all to do with privacy).
You will be hard pressed to find a host that will host pornography that is illegal in the EU (or the us) or actually hosting other peoples copyrighted material. (Not trackers or links by the way, their policy says the material itself).
I am not sure why anyone would want to help you. I am sure their are Slash dotters into kiddy porn, but it most be a very small number,
On the apple you just close the lid.. Not press the button, wait 5 seconds and close the lid. Most (I am sure there must be one that actually works somewhere) "PC" laptops will stay on if you just close the lid (even with sleep when lid is closed selected in windows) and then get really hot, overheat and shut down.
Than those editors accounts would be blocked....... This != Science of Rockets
People with reputations like the Editors of the Encyclopedia Britannica?m l
t he_Encyclop%C3%A6dia_Britannica_that_have_been_cor rected_in_Wikipedia
http://slashdot.org/articles/07/07/24/0114228.sht
http://en.wikipedia.org/wiki/Wikipedia:Errors_in_
Those are pretty funny since they are all just basic facts that were wrong.
The patents don't lock you into amazon any more than you may have been without the patents. If you forsake all software that has ever applied for a US software patent, your life might get a little dull. Public companies must make all reasonable attempts to protect their intellectual property, it is part of their fiduciary responsibility to their shareholders. Not applying for a patent on something that was developed internally at a publicly traded company is just irresponsible.
You also have no idea what Amazon S3 is. Please do not treat slashdot comments you have read as facts, go to Amazon and read the S3 FAQ. Web Service != Web Interface. Gdrive , thanks for the laugh.
It is not a web interface it is a Web Service. Webmin, NFS and AFS are all so entirely unrelated to any actual discussion of the S3 service it is not worth any further comment. RTFA then , RTFP, then go over to Amazon and RTFFAQ.
Wow thats a mess. Maybe it was because I sent it over EDGE, but I suspect it is more a function of my giant thumbs on the blackjack keyboard.
I have used sprint verizon EV-DO for over a year in about 25-30 US cities and canceled my service when I got my blackjack in Febrary and realized the tethered Blackjack on EDGE or 3G is quite a but faster than the EV-DO service. Theory is one thing and practical implementation is another entirely. At least in Dallas, Houston, San-Diego, Los Angeles, New York, Washington, Miami, Orlando, Chicago, Phoenix, Jackson, Denver and Philly the coverage, speed and reliability of the Sprint/Verizon network to be nearly as good as I receive on my phone with bluetooth PAN. I realize a lot of those markets are 3g, but 3g is generally limited to the core of the market and even in outlying areas like my house that only has EDGE service, the Cingular network performs considerably better. Slingplayer is unusable on EV-DO network and can work pretty well on the EDGE network. Perhaps if you are just doing something bursty like web traffic , you might be ok with EV-Do, might even think it is better , but for sustained bit rate I have never seen that to be the case.
I think most users will not see a huge difference between the network speeds. I have been using my blackjack since february. Tethered (bt pan) 3G generally tests out 280-350k range and EDGE in the 100-150k. Both have bursts and valleys (! have seen sling player on tethered pc jump over 600k for a bit on 3g. My house is edge only, it is fine for web browsing and even the mobile slingplayer. Many people seem to mistakenly think, cingulars edge network is the same as that terrible mess you get access to with a Verizon/sprint edge card. Tethered cingular is much faster and more reliable than anything ever experienced with one of those cards.
Battery life on 3g would be a huge problem for most consumers, the 3g drain is continuous when connected to a 3g network, not just when transfering data. 3G is just not ready for the mass market. I know we like to think everyone is like us, but the iphone was not made for the nerd community.
You are correct though.. A significant portion (likely somewhere between most and almost all) of the people on the Internet where someone is labeled as a fan boy are simply people trying to correct information that is factually incorrect. It does not mean the poster is mentally ill, a zealot, insecure or anything else. It simply means they were interested in injecting facts in a discussion on the Internet. Unfortunately for them, most people on the Internet are not interested in facts...
Please link us a real world example of a virtualization project where they experienced an actual increase of their electirc bill after the consolidation/virtualization was increased. I linked several that showed a significant decrease. You will not find any, because it does not happen. (Numbers you continue to just make up randomly in your head notwithstanding).
I can tell you 5 things about yourself:
1. You have never participated in a data center virtualization project of any size
2. You have never participated in a data center migration project of any size
3. You have never worked in a data center above the level of an untrained operator
4. You have never designed or evaluated any data center or significant server infrastructure.
5. You are incapable of admitting you are wrong.
Maybe if you can find the professional help you need to fix #5, you can learn how 1-4 actually work.
More reading you will ignore..I am sure if you actually read some of this , you would just stop posting..
Overall, consolidation with virtualization and blades can offer significant help with both space and power and cooling, but awareness of power/cooling requirements and careful planning are essential.
http://searchservervirtualization.techtarget.com/c olumnItem/0,294698,sid94_gci1243507,00.html
Upon the initial migration of the workload from 120 servers, AISO.net achieved a drastic 50-60 percent reduction in power consumption
http://www.mysiriuszone.com/index.php?option=com_d ocman&task=doc_download&gid=1708&Itemid=292
contain such virtualization features. According to Waugh, virtualization boosts overall performance while reducing server room power consumption.
http://www.serverwatch.com/hreviews/article.php/36 39556
It's tempting to say that after long periods of testing, an increasing number of organizations are seeing the benefits of running applications on virtual servers housed in fewer physical boxes: increased resource utilization, faster server implementation, fewer devices to manage, lower management costs, a smaller data center footprint, and lower power and cooling costs.
http://www.serverwatch.com/tutorials/article.php/3 637631
Processor virtualization allows concurrent operating system execution environments to co-exist and share a fixed set of hardware resources. One of its many advantages is that this technology facilitates server consolidation, reducing both operating costs and power consumption.
http://whitepapers.techrepublic.com.com/whitepaper .aspx?docid=289268
"Data center power demands are growing at an unsustainable rate," said Bill Zeitler, senior vice president, IBM Systems and Technology Group. "The most important thing now is utilization of servers through technologies like virtualization."
http://searchdatacenter.techtarget.com/originalCon te
I am going to respond because you are just a moron! Not total per rack TOTAL! Lets say you have 100 servers and they are in 10 RACKS. You consolidate these 100 Servers in 10 racks through virtualization. You now have 10 servers in 1 rack, carrying the load of what was 100 servers in 10 racks. The total power for that 1 rack will go up! The total power overall will go down, EVERY SINGLE TIME. You win the award for the densest person on the Internet. I have to assume you are just fucking with me at this point because no one who can type can be as idiotic as you. (I gave you an out , you might want to take it).
s avings-through-virtualization/
y s_reward.html
d ation.php
t ry=entry070517-222751
AC Costs absolutely become different! Some systems can not handle the higher density BTU load properly! But who cares because your assertion all along is that total power consumption goes up, not "Your current cooling system might need to be redesigned to accommodate the denser heat output". If you would like to discuss the impact of virtualization on Data Center cooling , I would be happy to. First you need to let us all know that you have finally grasped the fact that total power consumption does not go up. Until you can grasp that fact there is no point in trying to help you understand anything else.
The fact you think you need an additional cooling system proves you do not understand data center cooling either, 9/10 times the problem can be addressed with proper planning of the rack layout and its relationship to the existing CRAC units. Want to know why? Because Total BTUs went down right along with the Total Power. You of course do not believe using less power overall actually is a reduction in total power consumption. I am sure you will now link me some article you misread that you believe states virtualization increases the total BTUs in the datacenter.
You are just grasping at straws, trying to save some face. Man up, admit you were wrong and move on.
You like to read stuff on the Internet. You seem to focus on writers speaking hypothetically. Here are some real people talking about their real power and btu reductions:
http://virtualize.wordpress.com/2006/12/18/power-
http://blogs.hds.com/david/2006/11/a_power_compan
http://www.foedus.com/solutions/datacenterConsoli
(conslidation and virtualization are 2 different ways of skinning the same cat, you still end up with fewer servers at a higher load)
http://www.solutioncentre.co.uk/blog/index.php?en
You are showing your lack of first hand knowledge on the subject. Although many data centers still have 2 20-amp circuits per rack, you will find almost none built out in the last 5-10 years with less than 2-30amp circuits for racks and power distribution is generally built much more flexibly these days to support virtualization, blade servers and other high density configurations. You need to stop getting all your information from Internet articles. They are generally written by writers and not experts in the Technology they are writing about. Anyone who built a new data center today only designed to handled 2-20amp circuits per rack would be fired.
The second article you linked also say "a physical server" nothing about increased overall power consumption.
One of the first problems often overlooked is the high power consumption and high heat output of a physical server hosting multiple virtualized servers. Remember that virtualized servers cause the hardware to run at a higher utilization rate, which means the power required to operate the physical server hardware goes up, as does the heat output.
A physical server running many virtual machines may run at a constant 80% or higher utilization rate, at which point the hardware is pulling quite a bit more power than a non-virtualized server running on average around 15% utilization.
The point is of course that many data centers are not designed to handle the increased power in the smaller footprint and may not be equipped to deal with the increased heat. It is all right there if your are able to read and comprehend. No where does it state overall power consumption will be increased. I am not going to discuss this with you any more because it is very clear you are simply incapable of admitting you have no idea at all what you are talking about. All you do is continue to link and misquote articles you did not understand to begin with.
The only appropriate response from you at this point is "I am sorry, I misread both of the articles I linked" anything else is just going to make you look like more of an ass than you already do. It is not a situation where you can agree to disagree or say it is a matter of opinion. You are just wrong.
You are clearly not able to understand what "a physical server" is. Older data centers may not be equipped to handle the increased power/btu per square foot of virtualization (or blades for that matter). This is the hidden cost your last article refers to, it has nothing at all to do with increased overall power consumption.
I can see where someone with no background in data center technology might fail to make the distinction, thats ok I am sure you are good at whatever it is you actually do.
If there is any more of this discussion you are talking to yourself, it is impossible to teach someone who believes they know everything.
You know all the author was trying to point out is that you should not base your projected savings on the 20W number, I am not sure why you think it meant something else.
At this point it would be appropriate to interject, "all other things being equal"--meaning that there is no difference in application performance--a big assumption, and that the identical servers are used before and after
Type of servers used before and after = irrelevant. All things being equal (As I stated) is about application load not changing. because that has nothing to do with the virtualization process.
I am starting to feel sorry for you now.
I can tell you after virtualizing thousands of servers I have never seen a case where total power consumption has gone up. Working with IT personnel in dozens of large companies involved in virtualization projects, I have never heard even one suggest it might be happening to them.
You seem to believe a lot of things to be true, but none of them are based on anything you have actually done. I am sure this is the point where you claim I am full of crap. It does not matter if you believe me or not, because the facts speak for themselves and all you have to stand on is a really poorly written article by someone only slightly better informed than you.
Except 5 has nothing to do with what he said, you can't take my 5 and his 20 and 280 and think you have solved a riddle. You really are just embarrassing yourself now. What you just claimed would happen is not what happens in reality , period. There is no place in that article where the author tells you there will be an overall increase in power used. Period, end of discussion. Do you know why there is no place in that article where it says that? Because it would not happen. At no point does the author say total consumption before virtualization was 100W, you made that up based on a completely unrelated example I made.
One thing I can tell you based on the information provided in the article. Total power used before virtualization was more than 280W . Show me a direct quote from the article where it says total power used before virtualization was less than 280W. (You won't find it). I am really sorry I have proved to you multiple times that you misunderstood (and continue to misunderstand) the article. You obviously know absolutely nothing about this subject, I would stop replying if watching you make silly statements was not so dang entertaining. Mark Twain would be happy to see you still proving his point so many years later.
Actually the fact they are not linear is exactly why you are wrong. The overhead in operating additional machines is exactly the difference. You should really stop arguing about something you clearly do not understand. It was obvious you did not understand the article or what he meant by footprint. To call me foolish for pointing out your own misunderstanding only serves to prove my point.
To clarify what he is saying for you..He is saying the power per "FOOTPRINT" is going to go up, not the power per OS Instance. That is a no brainer, but he is being extremely vague in order to attempt to make a point. The power per FOOTPRINT is the power per Physical box. Wow! It is going up? Absolutely!
The number of physical boxes is going down. The author (Who apparently also has no experience actually implementing this) either doesn't get it, or was intentionally misleading to try and make a sensational point.
Reducing physical servers through virtualization will always reduce overall power consumption. You can continue to try to deny it all you want, but you really are just misunderstanding the article.
Somehow I doubt you will post here and tell us all that you did in fact misunderstand what he meant by footprint.