I think about this question a lot, and it is one I am asking more as well to customers and peers in the industry. There are hundreds of facilities that were built 20-30 years ago and a few dozen that have been built in the past 10 (remember the .com bubble?) and I am being conservative. They were built with a solid approach for 20 years ago - build a building using a watts/ft metric to calculate costs, density, and ultimately capacity. Remember that Wang, Digital/DEC, and IBM mainframes were the systems of choice. Then the PC came on the scene, and drastically changed the dynamic of computing. Client/Server, new network protocols, ARPA, TCP/IP and http - the evolution was on.
So too were manufacturers evolving - building smaller footprint, higher capacity machines. These machines were 2-3x as powerful as their predecessors, taking up half the space. Companies could consolidate their huge footprint monlithic systems inside their data centers and make room for more. Or could they?
Today, to me it seems like deja vu all over again.
The computing world has shifted again by using virtualization, and those smaller more powerful servers, just got smaller (so small you can't touch or see them), went virtual, and companies are embracing the next evolution with gusto. They are also hitting the wall that IT managers who are retired or dead hit decades ago - power.
Why is this happening again? In a word - heat.
These smaller machines draw more power, and generate a lot more heat for their footprint. Ask a multi tenant data center operator who services the retail market on a cabinet by cabinet basis and they will tell you that cabinets full of gear that draw 4.8Kw are going in next to cabinets with 10-14 Kw. The higher density cabinets generate a lot more heat that needs to be removed from the floor so all the densities of the servers with their heat signatures can be maintained properly.
What is intrinsic of a 20 year old facility that makes it better to satify growing densities, evolving computing applications and hardware, and whatever comes down the pike? How does the colocation market evolve at the same time? When do we hit the wall, or do we hit the wall? Can you simply throw more air conditioners/handlers on the floor? Can we just continue to push utilities to increase generation and distribution? What happoens to the cost models?
I am starting to think that we are at a crossroads. Why do I believe this? Because data centers are starting to evolve. Containers, modular buildings, more flexible options. Until the hardware manufacturers get together with data center operators and do some R & D in a meaningful way, share roadmap, do testing together, we will continue to build the way we have, and not the way we must.
Want to play together - ping me. I want to figure this out...
Showing posts with label high density hosting. Show all posts
Showing posts with label high density hosting. Show all posts
Tuesday, August 17, 2010
Tuesday, February 3, 2009
Why High Density isn't Dense Enough
I have been involved in four discussions in as many days about High Density Hosting. When I ask - what does that mean to you? I usually get a number that is north of 125 watts a square foot up to 200 watts a square foot.
That is one way to look at it, but even at 200 watts a square foot, if your loading up a multi core box with dozens of VM's you will be headed north of that. Typically what happens is that hosting companies balk at the space required to cool 200 watts plus. Looking at power density is about power NOT space.
So in one conversation the company said that they wanted to deploy 10 - 30Kw cabinets. Total draw of 300Kw.If I throw in a PUE of 1.5 I am at 450 Kw for draw. If I have a 10,000 square foot pod that doesn't have a half a megawatt available in either current or generator power, it can be the best space on earth but you can't power it. No power = no CPU cycles, and no company.
Buy power, not space. Think about when the lights go out - you aren't thinking about what a nice neighborhood you live in or what a good investment it was, or how sweet the location is. You just want to nuke some popcorn and watch the Red Sox beat the Yankees.
That is one way to look at it, but even at 200 watts a square foot, if your loading up a multi core box with dozens of VM's you will be headed north of that. Typically what happens is that hosting companies balk at the space required to cool 200 watts plus. Looking at power density is about power NOT space.
So in one conversation the company said that they wanted to deploy 10 - 30Kw cabinets. Total draw of 300Kw.If I throw in a PUE of 1.5 I am at 450 Kw for draw. If I have a 10,000 square foot pod that doesn't have a half a megawatt available in either current or generator power, it can be the best space on earth but you can't power it. No power = no CPU cycles, and no company.
Buy power, not space. Think about when the lights go out - you aren't thinking about what a nice neighborhood you live in or what a good investment it was, or how sweet the location is. You just want to nuke some popcorn and watch the Red Sox beat the Yankees.
Thursday, May 15, 2008
Virtualization – The chargeback conundrum
Funding it vs. Paying for it...
I have recently had 6 discussions in the past two weeks with companies on how to get a Virtualization project funded, which is a relatively simple discussion to have. It basically involves us working with our ROI model to come up with real cost savings on quantifiable (hard) costs. Based on that the projects are either funded or not and we help roll out infrastructure (on premise or a third party site), licensing, etc. – the usual project lifecycle stuff.
Where clients fall down in on how to get business units and departments to pay for what they use. I have dubbed it the Chargeback Conundrum.
The issue is that when an enterprise wants to execute an Enterprise License Agreement (an ELA) for the Virtualization software (VMWare) there are no simple controls or visibility into what businesses consume and how to charge them back. From the business perspective, they are reluctant to go all in to help pay for it when there is no proof of what they are getting or understanding of what they will be charged, and arguably more important – no trust with the IT organization.
In other words –
Business Executive: How do I know that I am not giving a piece of my budget to another group?
IT Executive: (insert explanation here).
Business Executive: OK, prove it.
IT Executive: (insert deer in the headlights look on face here)
If you need help – reach out – inquiries@virtualizationstuff.com
We can give you unprecedented visibility as product or a service so if you don’t want the cap ex on your balance sheet we can accommodate.
The services offering for Virtualization is also preferred when you want to pay for the functionality vs. be responsible for it.
You’ll need to work with your facilities folks to look at power consumption. While you can shrink the footprint, the power per cabinet jumps from 4KW to 7KW (almost doubles), and you’ll want to have the specialized floor tiles with variable speed fans, the temperature sensors on the front and back of the cabinets with the blades in them, and have it tied together in a management console.
If it sounds too cumbersome – reach out. We’re great at cutting to the chase…
inquiries@virtualizationstuff.com
I have recently had 6 discussions in the past two weeks with companies on how to get a Virtualization project funded, which is a relatively simple discussion to have. It basically involves us working with our ROI model to come up with real cost savings on quantifiable (hard) costs. Based on that the projects are either funded or not and we help roll out infrastructure (on premise or a third party site), licensing, etc. – the usual project lifecycle stuff.
Where clients fall down in on how to get business units and departments to pay for what they use. I have dubbed it the Chargeback Conundrum.
The issue is that when an enterprise wants to execute an Enterprise License Agreement (an ELA) for the Virtualization software (VMWare) there are no simple controls or visibility into what businesses consume and how to charge them back. From the business perspective, they are reluctant to go all in to help pay for it when there is no proof of what they are getting or understanding of what they will be charged, and arguably more important – no trust with the IT organization.
In other words –
Business Executive: How do I know that I am not giving a piece of my budget to another group?
IT Executive: (insert explanation here).
Business Executive: OK, prove it.
IT Executive: (insert deer in the headlights look on face here)
If you need help – reach out – inquiries@virtualizationstuff.com
We can give you unprecedented visibility as product or a service so if you don’t want the cap ex on your balance sheet we can accommodate.
The services offering for Virtualization is also preferred when you want to pay for the functionality vs. be responsible for it.
You’ll need to work with your facilities folks to look at power consumption. While you can shrink the footprint, the power per cabinet jumps from 4KW to 7KW (almost doubles), and you’ll want to have the specialized floor tiles with variable speed fans, the temperature sensors on the front and back of the cabinets with the blades in them, and have it tied together in a management console.
If it sounds too cumbersome – reach out. We’re great at cutting to the chase…
inquiries@virtualizationstuff.com
Friday, January 11, 2008
Virtualization Hosting Dependencies
I mentioned that there were a few dependencies around virtualized hosting which in my opinion are not that dramatic, unless of course you own a data center that needs to be refurbished...
Power is the biggie, with cooling right up there as well. Most cabinets/racks draw 4KW a rack (42U cabinet) if you through a loaded Blade rack in the data center, you are looking at 7KW, essentially double. So what, right?
You will need to have the ability to do high density hosting which means 200W a square foot vs. the standard 100W a square foot.
For cooling you will need/want special tiles in the floor that increase airflow around the cabinets. For safety's sake you'll probably want to drop another CRAC in for sites with seasonal spike in utilization and corresponding heat signature(s).
If you want a green angle on the cooling, look North my friends. The smart data centers are using mother nature's inherent cooling capabilities to keep gear cool. They shut down several CRACs and use an exchanger to draw very cold dry air into the facilities, reducing the carbon footprint of that data center through reductions in power usage and cooling.
Depending upon the age of data centers this will vary from extremely difficult/expensive/not worth it to a few weeks of refitting.
Power is the biggie, with cooling right up there as well. Most cabinets/racks draw 4KW a rack (42U cabinet) if you through a loaded Blade rack in the data center, you are looking at 7KW, essentially double. So what, right?
You will need to have the ability to do high density hosting which means 200W a square foot vs. the standard 100W a square foot.
For cooling you will need/want special tiles in the floor that increase airflow around the cabinets. For safety's sake you'll probably want to drop another CRAC in for sites with seasonal spike in utilization and corresponding heat signature(s).
If you want a green angle on the cooling, look North my friends. The smart data centers are using mother nature's inherent cooling capabilities to keep gear cool. They shut down several CRACs and use an exchanger to draw very cold dry air into the facilities, reducing the carbon footprint of that data center through reductions in power usage and cooling.
Depending upon the age of data centers this will vary from extremely difficult/expensive/not worth it to a few weeks of refitting.
Subscribe to:
Posts (Atom)
Posts
