Slashdot Mirror

All fair criticisms, but it's a first step. The EPA plans to address many of the shortcomings of the current Energy Star for Servers program in an expanded Tier 2 spec that is scheduled to arrive in the fall of 2010. The update is intended to expand the program to include blade servers and servers with more than four processors.

IBM's water-cooling rear-door heat exchanger (known as "Cool Blue") was judged the most energy-efficient entry in a "chill-off" between cooling vendors last year. The contest was sponsored by the Silicon Valley Leaders Group and conducted in the data center on Sun's Santa Clara HQ campus. The same unit is also sold by Vette Corp., which offered a video demo at Data Center World.

IBM's water-cooling rear-door heat exchanger (known as "Cool Blue") was judged the most energy-efficient entry in a "chill-off" between cooling vendors last year. The contest was sponsored by the Silicon Valley Leaders Group and conducted in the data center on Sun's Santa Clara HQ campus. The same unit is also sold by Vette Corp., which offered a video demo at Data Center World.

With all the talk of container and "lego" data centers, Oracle wants to become fully vertically integrated so that you can go to Oracle and say: "I've got $10 million -- sell my data center blocks".

Sun's already been developing their own data-center-in-a-shipping-container, and Oracle now has all the bits and pieces:

• Hardware that runs Oracle really well -- Sun SPARC
• The operating system for big data centers -- Solaris
• The Java application server -- BEA's WebLogic
• The Database -- well duh!

Also, having a horde of hardware engineers is Ellison's wet dream. As I said before, Larry Ellison wakes up every morning and asks himself, "How can I [fsck] Microsoft today?" Larry has stated in the past he wouldn't mind moving beyond databases, and with Sun's hardware and Java, he's poised to do pretty much anything he wants. So he might entertain delusions of mobile, return of the net appliances, home multimedia, etc. In the short term, though, I think he's hoping he can create custom hardware to make Oracle and Java run much faster. Will he succeed? Dunno, but Larry Ellison has a ferocious desire to succeed, and often, that's all you need.

There are plenty of steps that should be prioritized over soy-based ink:

• Your server room can run at a hotter temperature, without increasing failure rates. Set it to the max of what is comfortable to work in.
• You can probably virtualize quite a physical few servers out of existence.
• When/if your offices are air conditioned, make sure you use energy-efficient lighting, turn off workstations overnight etc etc.
• Make sure there are enough bike-racks outside the office.
• Provide a shower for those that want to bike/run/rollerblade to the office.
• Make sure the office heating system adjusts temperature overnight.

Once these steps are done (the company will profit from most of them), feel free to consider soy-based toner cartridges.

This is a project they should get right. Social Security's data centers maintain earning and benefits information for nearly every American worker, processing 75 million transactions per day. "All of our plans depend upon a strong, 21st century data center," the agency said in a recent document describing its needs. See Data Center Knowledge for more on this. Regarding the location, the agency says the new facility would need to be within 40 miles of its current location due to "data linkages" - presumably a reference to the ability to perform real-time active-active backup, where distance limitations of storage networking protocols come into play.

If you are a business trusting mission critical data or computing to the cloud, then you need to verify how well the cloud handles these issues. Just because they have a big data center doesn't mean that they have redundant services spread across the data center and multiple data centers in case one is hit by a disaster. http://www.datacenterknowledge.com/archives/2009/03/23/carbonite-lawsuit-reveals-data-loss/

That bunker, that one of their ISP has may just come in handy.

Actually, that ISP is one of our best supporters and we have all of our Pirate Party servers in that bunker. ;-)

so what we have here is a possibility that in the future a 'pirate' party controls the government maybe? Would Obama with his RIAA lawyer friends declare Sweden to be part of axis of evil and will actually bomb them to bring in the democracy US style (where only 2 parties are really allowed to hold the government in practice).

That bunker, that one of their ISP has may just come in handy.

http://www.datacenterknowledge.com/archives/2009/04/15/inside-the-james-bond-villain-data-center/ It seems like these folks have a lot going for them. There is the sensible management, a cool place to work, etc.

A neat video tour of the "james bond villain data center": http://www.datacenterknowledge.com/archives/2009/04/15/inside-the-james-bond-villain-data-center/

Well, these guys have a nuclear bunker for a data-center, they probably think that even if the government comes and attacks them, they can just ride it out inside. They'll probably survive even if US decides to blast them with a nuke (I wonder what the rest of the world would think of the USA if that happened though - US blasting an entrance into a datacenter with a thermo-nuclear weapon in a populated Swedish area. Oh well, just pretend there are WMDs in there and anything would go...

They take care of their customers and can still run after a nuclear war. (and you know some guy in there is doing the maniacal laugh every once in a while) http://www.datacenterknowledge.com/archives/2009/04/15/inside-the-james-bond-villain-data-center/

Remember those Slashdot articles a few years ago about Google buying up dark fiber?

And more recently, building massive data centers near power stations?

I wonder if they might be waiting for something like this to open up their ISP division and bury Comcast and TWC by offering unmetered service?

About how much has Facebook saved by using Open Source Software? I ask because I am not familiar with licensing costs from competing solutions. Thanks!

I haven't watched the presentation so don't know if this is answered there, but it's hard to pin down any numbers on precisely how many servers facebook operates. That said, an estimate of their expected power usage in their recently acquired second datacenter is 6 megawatts, placed at twice the usage in their current datacenter. Realistically, this probably equates to a cluster of around 5,000 machines in the current datacenter.

Costs per machine are likely to be restricted to Windows Server Web Edition; other software would not be needed on all machines (depending on cluster architecture, of course) so would be a trivial cost in comparison. Retail for the web edition is $399; I think we could expect such a high profile user to qualify for a 50% discount. This would put their software costs at about $1M. Considering that they're believed to have spent over 100 times this on hardware and support costs over the last year, I doubt this would be a particular concern. Price of purchase is not a factor in why facebook does not run on proprietary software.

Fair point -- I never thought of that. However, there are some other links to this:

• Inside a Google Data Center, including a video
• Notes from someone who says they were at the conference (also aggregated at the ACM

Maybe this is legit...

Date Center Knowledge has videos of the secret server and a tour of one of the container data centers.

Date Center Knowledge has videos of the secret server and a tour of one of the container data centers.

What exactly is this QZHTTP?

I honestly don't know. Never heard of it before now, my Google Fu finds nothing in English. Indicating it is most likely propriety to Tancent QQ ...

I hope this didn't affect the IPv4 exhaustion date.

I guess this could also just be a whole lot of fuss over something that will become common place. I mean with the event of virtualization, hilarious 32 core chips due out and predictably cheap storage/memory ... won't every large company soon be able to foot the bill on and house (what appears to be) 20 million web servers? I guess IP addressing, routing & bandwidth will always be a problem but the hardware is sure getting to the point.

This incident knocked several major hosting providers offline, including Media Temple in Los Angeles and Canada's iWeb.

Actually, some folks are already considering the possibility.

The Scottish tidal power project actually involves two huge data center projects. Atlantis is working with Morgan Stanley on a large data center near its planned tidal power generation site in the Pentland Firth. Internet Villages is planning a large data center campus near Dumfries that could eventually include 3 million SF of data center facilities. The alliance will split the responsibiltiies, with Atlantis handling power generation at its own site and a second location nearer the Dumfires project, while IVI will handles the marketing and be responsible for finding customers for both facilities.

The Scottish tidal power project actually involves two huge data center projects. Atlantis is working with Morgan Stanley on a large data center near its planned tidal power generation site in the Pentland Firth. Internet Villages is planning a large data center campus near Dumfries that could eventually include 3 million SF of data center facilities. The alliance will split the responsibiltiies, with Atlantis handling power generation at its own site and a second location nearer the Dumfires project, while IVI will handles the marketing and be responsible for finding customers for both facilities.

It's client-server all over again? Umm. Yeah? So?

So, what happens when you can't get to the server because there has been a fire in your data center? Or, a crucial fiber optic center? Or, a meteor could be involved.
What do you do when your server, or your customer's service, is unavailable for a week or three?

For the services that they provide...

Wikipedia now has 200 application servers, 20 database servers and 70 servers dedicated to Squid cache servers. Reference

I'd say that it is quite remarkable.

There are a number of companies providing commercial DC solutions for data centers. Validus DC Power is providing products for DC power distribution, while Power Loft is building a brand new data center optimized for DC power.

There are a number of companies providing commercial DC solutions for data centers. Validus DC Power is providing products for DC power distribution, while Power Loft is building a brand new data center optimized for DC power.

They are trying to fix it, much the same way that Sarbanes-Oxley fixes accounting problems. Communications providers are required to keep an 8 hour power backup on all sites and 24+ backup on 'important' sites like switching centers or something along those lines. The idea is that storms like those that hit New Orleans would not cause the problems that they did. This storm is exactly the thing this measure by the FCC is supposed to fix... HAHAHAHAHAHAHAHAHAHAHA

Legislators (Goodbye FCC) do NOT know how to run businesses. Some perhaps, but on the whole they are terrible business advisers and this legislation only proves it in the aftermath of this storm. I hold a harsh opinion of this situation because AT&T should have had backups in place to handle this situation. All Communications providers deal with such things and AT&T has enough history to know what to do... shame on them.

This will only work if they make un-metered applications illegal and move OS functionality to the cloud and of course patent cloud computing ..

http://www.datacenterknowledge.com/archives/2008/02/07/cable-cut-fever-and-the-fifth-cut/

There weren't five cuts. This was known 10 months ago. You seem to be selective about what you read.

I don't buy the original explanation that 2 ships were able to cut 5 cables in different locations.

One of the cables near Egypt that was cut had video footage and it showed no ships at the time it was cut.
http://www.datacenterknowledge.com/archives/2008/02/04/egypt-ships-didnt-cut-internet-cable/

Well, the artist's rendition sure dosen't make it look very safe.

It looks as sturdy as Windows XP's default wallpaper, they might as well paint a big bullseye on it, its inhabitants should hope that the neighbor's kid dosen't own a BB gun or that the tree dosen't come crashing down on it, etc. etc.

yeah and I thought Texas was a center for data centers? (ofc not saying they're all running linux) http://www.datacenterknowledge.com/archives/2007/08/16/the-texas-data-center-phenomenon/

What the article doesn't say is that the NSA leased the former Sony site in 2005, with plans to locate as many as 6,000 employees at the site. It had scaled back those plans in 2006, but then decided to build a data center on the property in 2007. Microsoft's project certainly confirmed San Antonio's qualities as a data center site, but it's not really a case of the NSA following Microsoft.

The summary misses a key point: Sprint has restored its connection to Cogent, meaning the two companies can pursue their lawsuit and grievances without using customers as bargaining chips.

Google said recently that it runs its data centers at 80 degrees as an energy-saving strategy, so chips that support higher temperatures would mean fewer hardware failures in their data center. Most data centers operate in a temperature range between 68 and 72 degrees, and I've been in some that are as cold as 55 degrees. Lots of companies are rethinking this. In the Intel study on air-side economizers, they cooled the data center with air as warm as 90 degrees. ASHRAE is also considering a broader temperature range for data center equipment.

For a list of interesting places to locate a data center, see data centers in strange places/

For a list of interesting places to locate a data center, see data centers in strange places/

Lots of interesting comments on this thread. The whole issue of heat and humidity levels in the data center was the subject of a heated debate last fall at Data Center World. Mark Monroe from Sun suggested raising the set point from 68 to 72 degrees and got all kinds of pushback from the audience.

A key concern about higher set points is that nudging the thermostat higher can save money, but may leave less time to recover from a cooling failure. That would certainly be relevant if you're using free cooling when it's 90 degrees outside.

Data Center Knowledge has a video in which the Intel engineers who conducted the study talk in detail about the setup and the results.

Sun is also running a comparable experiment with Belgacom and allows you to log in to a live interface to view stats on in- and outlet temperatures and more at http://wikis.sun.com/display/freeaircooling/Free+Air+Cooling+Proof+of+Concept For more details and analysis see http://www.datacenterknowledge.com/archives/2008/09/18/intel-servers-do-fine-with-outside-air/ or http://securityandthe.net/2008/09/18/intel-sees-the-future-of-datacenters-and-it-does-not-include-airconditioning/

DC Knowledge also has a nice video of this experiment at http://www.datacenterknowledge.com/archives/2008/09/18/video-intels-air-side-economization-test/

Sun is also running a comparable experiment with Belgacom and allows you to log in to a live interface to view stats on in- and outlet temperatures and more at http://wikis.sun.com/display/freeaircooling/Free+Air+Cooling+Proof+of+Concept For more details and analysis see http://www.datacenterknowledge.com/archives/2008/09/18/intel-servers-do-fine-with-outside-air/ or http://securityandthe.net/2008/09/18/intel-sees-the-future-of-datacenters-and-it-does-not-include-airconditioning/

DC Knowledge also has a nice video of this experiment at http://www.datacenterknowledge.com/archives/2008/09/18/video-intels-air-side-economization-test/

Could the Google data barges survive a Hurricane? They don't look like they could outrun one, but Google's application says some of the Pelamis units could be submerged enough to ride out a major storm.

The article briefly mentions a survey on cloud computing released today by Pew Internet, which warns that "sloud users show high levels of concern when presented with scenarios in which companies might use their data for purposes users may or may not fully understand ahead of time. This suggests user worry over control of the information they store online." That includes using personal information for ad targeting.

Earlier this year Princeton University held a forum on cloud computing, which included an in-depth session of data ownership in the cloud and the issues it raises. It's available on YouTube in its 90-minute entirety.

The patent application confirms that Google has data center containers. The company patented the concept several years ago, but some developers of the Google container have said they abandoned the project. But the filing describes containers being lifted on and off ships by cranes.

All the really good evil is coded and compiled deep inside the top-secret Google data centers, surrounded by moats filled with sharks with friggin' laser beams on their heads.

Conspiracy theories aside, the data centers are a major innovation, and an area where Google has set standards for its competitors to chase. Google's massively scalable infrastructure is a big part of what has set it apart.

Looking at the cooling issue more broadly, here are a couple of resources that provide good information on techniques to optimize cooling on a budget:

The Hot Aisle: Great blog from Steve O'Donnell with practical ways to implement hot-aisle or cold-aisle containment and economizers.

Data Center Knowledge: Recent articles look at "roll your own" thermal monitoring solutions and using excess heat in swimming pools and greenhouses.

True, but Washington State recently withdrew a tax break for data centers, ruling that they're not eligible for a tax incentive for manufacturers because they don't actually make anything. I think the savings from the power benefit are larger than the tax cut, but some companies want both. Iowa's ready to deal, and it's not alone.

You might be overestimating the construction jobs, since Microsoft is experimenting with server farms in shipping containers, like those Sun Modular Datacenters. Des Moines already has massive IT infrastructure since it's the world's biggest centers for insurance data processing. They're going to use infrastructure that is already there.
No, I don't see many local jobs coming out of this. I see more strain on our local infrastructure. I see more power consumption and more pollution from coal-fired power plants.

Three separate PCs, each running the same software on the same data, and if one gives a different answer, the entire machine gets taken offline and support paged.

The difference is, three PCs can be had for less than three thousand dollars, new, even with monitors and such. How much will one mainframe cost you?

Take the cost of the system and divide it by the cost of one minute of unscheduled downtime. There are environments where thirty unscheduled seconds a year is unacceptable. If one of your three PCs goes down, how long will it take the other two to get their act together, back out the mistake, and reply, correctly, to the request the third PC was handling? You don't know, do you?

By the way, a PC that costs a thousand dollars (monitor included) was designed to be thrown away. And machines in data centers don't have monitors. Statements like that make it sound like you've stood on a raised floor.

How about a design that lets you run applications 24 hours a day, 7 days a week, with no downtime required for system upgrades?

Does Google qualify? How about Amazon?

If Google loses an app server and a thousand queries hang, a thousand people shrug their shoulders and hit Ctrl-R to try again. If a Google DB server scribbles the database and crashes, Google will use a day-old copy while they resync. This won't cut it when a database that's a few seconds out of date means millions of dollars are unaccounted for.

There are areas where mainframes eat Unix systems for lunch.

Only if there's an irrational need for it to be exactly one machine.

There's hardly ever just one. Companies that need mainframes usually need several.

And here's something else you don't understand about mainframes: There are environments where your combined internal throughput should be in the terabyte-per-second range, maybe because once a year you have to buzz through tens of millions of extraordinarily complex tax returns, apply rule sets with over a century of changes in them, flag suspicious activity, reconcile them against bank and employer data you've been tabulating throughout the year, and generate a check or electronic deposit for most of them. And every day you're late is a day where you've kept two hundred and twenty five billion dollars in tax refunds out of circulation.

You gonna pull that plow with a team of oxes? Or are you gonna settle for ten thousand mice who spend half their day shouting "you still there?" at each other?

What happens if, say, that building explodes?

Funny you should ask. Long before you were born, many very serious adults dedicated a sizable portion of their careers to figuring out exactly what to do when this sort of once-per-country-per-decade disaster happens. Adults serious enough to consult geologists about seismic activity before they build their data centers. Serious enough that they receive regular reports on political stability in the countries that host their data centers. And bright enough to ask, "hey, what if we put some of our computers in a different building?"

I'll add a disclaimer, too: I work on a project which is currently deployed via Amazon EC2.

When they went down for two days last October, how much data did you lose? How much much damage was done to your reputation and revenue stream when they went down for two days in February?

Here's my disclaimer: I work for a company that processes