Facebook Open Sources Its Servers and Data Centers. Facebook has shared many details of its new server and data center design on Building Efficient Data Centers with the Open Compute Project article and project. Open Compute Project effort will bring this web scale computing to the masses. The new data center is designed for AMD and Intel and the x86 architecture.
You might ask Why Facebook open-sourced its datacenters? The answer is that Facebook has opened up a whole new front in its war with Google over top technical talent and ad dollars. “By releasing Open Compute Project technologies as open hardware,” Facebook writes, “our goal is to develop servers and data centers following the model traditionally associated with open source software projects. Our first step is releasing the specifications and mechanical drawings. The second step is working with the community to improve them.”
By the by this data center approach has some similarities to Google data center designs, at least to details they have published. Despite Google’s professed love for all things open, details of its massive data centers have always been a closely guarded secret. Google usually talks about its servers once they’re obsolete.
Open Compute Project is not the first open source server hardware project. How to build cheap cloud storage article shows another interesting project.
Can Open Hardware Transform the Data Center?
http://hardware.slashdot.org/story/11/10/30/0614234/can-open-hardware-transform-the-data-center
Is the data center industry on the verge of a revolution in which open source hardware designs transform the process of designing and building these facilities? This week the Open Compute Project gained momentum and structure, forming a foundation as it touted participation from IT heavyweights Intel, Dell, Amazon, Facebook, Red Hat and Rackspace.
Can Open Hardware Transform the Data Center?
http://www.datacenterknowledge.com/archives/2011/10/28/can-open-hardware-transform-the-data-center/
Is the data center industry on the verge of a revolution in which open source hardware designs transform the process of designing and building data centers? The Open Compute Project, an initiative begun in April by Facebook, is gaining partners, momentum and structure. Yesterday it unveiled a new foundation and board to shepherd the burgeoning movement.
If the project doesn’t succeed, it won’t be for lack of support. Yesterday’s second Open Compute Summit in New York featured appearances from executives for some of the sector’s leading names – Intel, Dell, Amazon, Facebook, Red Hat and Goldman Sachs. The audience was filled with data center thought leaders from Google, Microsoft, Rackspace and many other companies with large data center operations.
There are signs that the Open Compute designs could become more practical for a broader array of data center customers in the future.
Missing from the dais were companies specializing in power, cooling and mechanical design – areas where Open Compute designs are being shared.
Here is an interesting article that shows details of “Open Compute” Server and compares it to a HP server
Facebook’s “Open Compute” Server tested
http://www.anandtech.com/show/4958/facebooks-open-compute-server-tested
The Facebook Open Compute server design is ambitious: “The result is a data center full of vanity free servers which is 38% more efficient and 24% less expensive to build and run than other state-of-the-art data centers.” Even better is that Facebook Engineering sent two of these Open Compute servers to our lab for testing, allowing us to see how these servers compare to other solutions in the market.
The Facebook Open Compute servers have made quite an impression on us. Remember, this is Facebook’s first attempt to build a cloud server!
Comments:
It seems to me that the HP server is doing as well as the Facebook ones. Considering it has more featuers (remote management, integrated graphics) and a “common” PSU.
HP: Hard drive shortages hitting Google, Facebook DIY servers
http://www.zdnet.com/blog/btl/hp-hard-drive-shortages-hitting-google-facebook-diy-servers/64026
The do-it-yourself server crowd is apparently having trouble procuring hard drives due to the floods in Thailand.
Google and Facebook may already be squeezed by hard drive shortages, says HP’s CEO.
Facebook s new energy efficient data center
http://www.youtube.com/watch?v=a-xI4w0eR0Y&feature=related
Inside Facebook’s Server Room
http://www.youtube.com/watch?v=nhOo1ZtrH8c&feature=related
data center facebook
http://www.youtube.com/watch?v=-DRxqHrPrFw&feature=related
Facebook’s Oregon Data Center Uses As Much Power As Entire County
http://hardware.slashdot.org/story/12/01/31/0355228/facebooks-oregon-data-center-uses-as-much-power-as-entire-county
“The first phase of the Facebook data center in Oregon uses 28 megawatts of utility power, local officials said this week. That’s not extraordinary for a facility of that size in most data center hubs. But it stands out in Crook County, Oregon where all the homes and business other than Facebook use 30 megawatts of power.
[...] of the facilities and the thousands of computers that go inside. However, despite significant improvements in efficiency, the exponential growth in cloud computing far outstrips these energy [...]
Facebook unfriends 19-inch data center racks
http://www.theregister.co.uk/2012/05/02/open_compute_summit_open_rack/
Social media giant Facebook had built precisely one data center in its short life, the one in Prineville, Oregon, before it had had enough of an industry standard that was part of the railroad infrastructure and then the telephone infrastructure build outs and bubbles: The 19-inch rack for mounting electronic equipment.
At the Open Compute Summit in San Antonio, Texas, Frank Frankovsky, director of hardware design and supply chain at Facebook and one of the founders of the Open Compute Project, said it was time to get rid of the old 19-inch rack and give it a skosh more room – two inches to be precise – to do a better job of packing compute, storage, and networking gear in data centers and provide better airflow over components.
By stick with the current 19-inch racks and their limitations, “we all end up with racks gone bad,” explained Frankovsky
Given the tech industry experience with this sort of rack – and the fact that equipment rooms, tools, and techniques for bending metal to make electronics were all standardized on 19-inch racks, it made sense for the computer industry to adopt the same form factors when machines started getting racked up in volume in the late 1980s.
The rollout today of the Open Rack standard being proposed by the OCP is just one in what will no doubt be a lot of standards that are re-thought in the next several years by companies building hyperscale data centers.
The problem with 19-inch racks is that people try to cram too much gear into them, and they often end up poking out the back of the rack, which messes up the hot aisles in data centers. Maybe you can get a parts cart down the aisle now, and maybe you can’t. And racks are also packed with a bazillion network cables and heavy power cables, which makes maintenance of the machines in a rack difficult.
“Blades were a great promise,” said Frankovsky. “Companies needed help, and once blades came out, they said, ‘Please don’t help us anymore.’” By going back to the drawing board and coming up with the Open Rack design, Frankovsky says that the engineers at the social media company have come up with a scheme that is “blades done right.”
The OCP wants for vendors to adopt the Open Rack standard, but they may never do it except for the hyperscale data center customers that might pick up the Facebook designs and use them in their own data centers (or adapt the ideas with tweaks).
“Let’s face it, guys. There’s only so many different ways to bend metal,” said Frankovsky, referring to the ways that vendors try to tweak their rack designs to make them a little bit different and how they did not standardize (as they could have) on blade server and chassis form factors a decade ago. “By completely standardizing the mechanicals and electricals, this is going to help us stay away from racks gone bad.”
By going with a 21-inch wide rack design, the Open Rack can put five 3.5-inch drives across horizontally inside of a single server tray. Moreover, it can also put three skinny two-socket server nodes across on a tray and still have plenty of room for memory slots. Considering that the 3.5-inch disk is still the cheapest and densest storage device
The Open Rack also has power trays that are separate from the servers, which allows servers to be even denser and gives more flexibility in what you can put in the rack in terms of servers, storage, and networking. The idea is that when a new processor from Intel or AMD comes out, you replace as few of the components as possible to get the CPU upgrade and leave everything else in place.
Chinese web powerhouses Tencent, Baidu, and Alibaba were already working on their own custom rack designs that have some features similar to the Open Rack design, called “Project Scorpio,” and Frankovsky said in a blog post that the two camps were working out how to converge their respective racks to a single standard by 2013.
Building Efficient Data Centers with the Open Compute Project
http://www.facebook.com/note.php?note_id=10150144039563920
This meant we could:
Use a 480-volt electrical distribution system to reduce energy loss.
Remove anything in our servers that didn’t contribute to efficiency.
Reuse hot aisle air in winter to both heat the offices and the outside air flowing into the data center.
Eliminate the need for a central uninterruptible power supply.
The result is that our Prineville data center uses 38 percent less energy to do the same work as Facebook’s existing facilities, while costing 24 percent less.
Inspired by the model of open source software, we want to share the innovations in our data center for the entire industry to use and improve upon. Today we’re also announcing the formation of the Open Compute Project, an industry-wide initiative to share specifications and best practices for creating the most energy efficient and economical data centers.
Open Compute Developing Wider Rack Standard
http://hardware.slashdot.org/story/12/05/03/1417232/open-compute-developing-wider-rack-standard
“Are you ready for wider servers? The Open Compute Project today shared details on Open Rack, a new standard for hyperscale data centers, which will feature 21-inch server slots, rather than the traditional 19 inches. “We are ditching the 19-inch rack standard,”
Why Open Compute Is a Win For Rackspace
http://hardware.slashdot.org/story/12/05/24/2328249/why-open-compute-is-a-win-for-rackspace
“Cloud provider Rackspace is looking to the emerging open source hardware ecosystem to transform its data centers. The cloud provider spends $200 million a year on servers and storage, and sees the Open Compute Project as the key to reducing its costs on hardware design and operations.”
‘I think the OEMs were not very interested (in Open Compute) initially,’ said Rackspace COO Mark Roenigk. ‘But in the last six months they have become really focused.’
How Open Compute is a Win for Rackspace
http://www.datacenterknowledge.com/archives/2012/05/23/how-open-compute-is-a-win-for-rackspace/
Rackspace President Lew Moorman discusses the company’s support for the Open Compute Project during the project’s recent summit in San Antonio.
In the battle for the hyper-scale data center, long-dominant server OEMs like Dell and HP are doing battle with a growing challenge from firms offering custom server designs. If you’re looking for the front lines in this battle, look no farther than companies like Rackspace Hosting.
Rackspace is one of the fastest-growing cloud computing providers. The San Antonio company spent $202 million on servers and storage for customers over the past year, adding more than 12,000 servers in its data centers.
Rackspace is also keen on increasing its use of technology based on the Open Compute Project (OCP), which is developing efficient open source designs for servers, storage and data centers. The company believes that the OCP designs can help it reduce costs on hardware design and operations.
But who will build that hardware for Rackspace? That’s a multi-million dollar question that demonstrates the potential value of the Open Compute initiative for service providers and hyper-scale users. Rackspace currently has its servers customized by HP and Dell, but is also having conversations with original design manufacturers (ODMs) like Quanta and Wistron about their capabilities.
This growing vendor ecosystem creates options for Rackspace, according to Mark Roenigk, Chief Operating Officer of Rackspace, who says it has prompted incumbent OEMs to step up their game when it comes to Open Compute hardware.
“I think the OEMs were not very interested (in Open Compute) initially,” said Roenigk. “But in the last six months they have become really focused. I think we have a lot to do with that. We have Dell and HP engineers on-site now.”
When Rackspace launched its cloud computing operation, it had no supply chain beyond its relationships with Dell, HP and Synnex, Roenigk said. But that’s changed.
“Thus far we’ve taken bits and pieces of the Open Compute designs and incorporated them into our solution,” said Roenigk. ”Our hope and intention is that the Open Compute Project standardizes our design. We have no ambition to be in the hardware business. Open Compute can fill that gap.”
“Rackspace is totally committed to the Open Compute Project,” said Moorman. “We want efficiency and rapid innovation, and we can’t do it alone.”
And what about price savings on servers? While companies developing OCP designs like Quanta, Wistron and Hyve will be tough competitors, Roenigk says the key will be how HP and Dell deliver on their commitment to Open Compute. And price may not be the deciding factor.
“If HP and Dell can price something near where Quanta can, then I don’t have to call eight other guys to make it work,” said Roenigk. “Speed is important, because we’re growing so rapidly. In some ways, it’s more important than price. It’s a value proposition. There are times when I would rather get something now than get a better price later.”
“I was skeptical out of the gate,” he said. “I truly believe it’s real now. I think we’re seeing a fundamental shift I’ve not seen anything like in my 20 years in the business.”
Inside Facebook security: defending users from spammers, hackers, and ‘likejackers’
http://www.theverge.com/2012/5/25/2996321/inside-facebook-likejackers-spammers-hackers
If Facebook were a country, it would be the third largest in the world, just behind India and China. And like any country, Facebook has a police force to keep things under control. 300 people have been entrusted with the responsibility of keeping a 900-million-person virtual society from itself and from external forces.
Facebook’s deal with the world’s biggest anti-virus companies to include their blacklists in Facebook’s URL-scanning database got us thinking about other things the company does behind the scenes to keep its users safe, because a hacked, spammed, and depressed user isn’t coming back for more.
“Creating friction is the key to making users aware of what they’re actually doing,” Facebook Security and Safety team member Fred Wolens said, because a vast majority percent of “hacked” Facebook accounts don’t get hacked on Facebook.
Facebook starts by scanning the usual suspects of PasteBin-esque websites weekly to check for hackers dumping thousands of usernames and passwords. Facebook cross references credential dumps with its entire database of user credentials, then alerts any users that match to change their passwords.
Another measure Facebook takes is stripping every user of their referral URL when they click one of the two trillion links posted to Facebook every day.
A popular and nefarious way that spammers manipulate you is by putting invisible Like buttons on top of real buttons you can see like “Download File.” For example, if you’re trying to pirate an album from a suspicious site, the Download link might actually be a Like button that subscribes you to content from that site. Without even knowing it, you are liking a page and thus polluting your friends’s News Feeds with a spam post
Facebook responds to “likejacking” by sometimes showing a pop up that confirms whether or not you meant to Like that website.
The goal of many of these spammers is to generate impressions, just like banner ads do for content farm websites. Spammers get paid every time somebody clicks a link and sees an ad
When somebody has accidentally liked a page or clicked a nefarious link, it’s unlikely that their Facebook account will be compromised. The real problem is that most people use the same username and password on most sites they sign up for. When a user’s credentials for another site are stolen, thieves simply try them on banking sites and social networks like Facebook.
When someone friends you on Facebook, that request doesn’t always get through to your inbox. Facebook employs a complex algorithm to decide the likelihood that you know somebody, and whether or not to push through a friend request or file it as spam inside your “See All Friend Requests” folder.
Facebook’s database of malicious links contains billions of bad URLs, and its spam filters are precise enough that just .5 percent of users see spam on a given day, by its estimates
The difference for now is that we’re all choosing to use Facebook and explicitly accepting the company’s monitoring and control — they’re unfortunate preconditions of the virtual society. Without these rules, a site that entertains us for hours each day might descend into a spam and crap-filled cesspool, which isn’t very fun. And unlike the real world, if these rules change, it’s a lot easier to delete your Facebook profile than it is to relocate to another country.
Facebook smacks away hardness, sticks MySQL stash on flash
Replaces HDD with Fusion-io flashiness
http://www.theregister.co.uk/2012/05/31/facebook_fusion_io/
Facebook is using Fusion-io server flash cards in its datacentres to store MySQL data as well as process it faster because it’s better than using disk drives.
Piper Jaffray analyst Andrew Nowinski tells us that Facebook is using Fusion-io ioDrive PCIe flash cards for capacity as well as performance acceleration through flash caching
“The performance benefits of the Fusion-io solution greatly exceeded the HDD solution, making the price/performance much more attractive.”
Facebook is now “hosting [its] MySQL databases on Fusion-io cards, rather than using an HDD solution from LSI.” Nowinski tells us one reason is that MySQL has a journalling system wherein each row of data is written twice, once into the database table and again into the journal for redundancy and protection against a write failure. This uses up storage capacity and lowers performance.
He says: “By leveraging Fusion-io cards and the new SDK, customers can turn off this journalling system, since the Fusion-io cards also have a patented logging system that protects data in the event of power failure. By turning off journalling, management stated that customers can reduce the amount of data that is stored by 50 per cent, increase throughput by 33 per cent and decrease latency by 50 per cent versus the alternative HDD storage.”
Facebook likes wimpy cores, CPU subscriptions
http://www.eetimes.com/electronics-news/4375880/Facebook-likes-wimpy-cores–CPU-subscriptions
Facebook could start running–at least in part–on so-called wimpy server CPU cores by the second half of 2013. Long term, the company wants to move to a systems architecture that lets it upgrade CPUs independent of memory and networking components, buying processors on a subscription model.
The social networking giant will not reveal whether it will use ARM, MIPS-like or Atom-based CPUs first. But it does plan to adopt so-called wimpy cores over time to replace some of the workloads currently run on more traditional brawny cores such as Intel Xeon server processors.
“It will be a journey [for the wimpy cores] starting with less intensive CPU jobs [such as storage and I/O processing] and migrating to more CPU-intensive jobs,” said Frank Frankovsky
“We’re testing everything, and we don’t have any religion on processor architectures,” Frankovsy said.
Facebook published a white paper last year reporting on its tests that showed the MIPS-like Tilera multicore processors provided a 24 percent benefit in performance per watt per dollar when running memached jobs. Tilera is “the furthest along” of all the offerings, and they are “production ready today” with 64-bit support, he said.
Frankovsky noted several ARM SoCs and alternatives from both Intel and AMD are also “in the hunt.”
A rare look inside Facebook’s Oregon data center [photos, video]
http://gigaom.com/cleantech/a-rare-look-inside-facebooks-oregon-data-center-photos-video/
In a rare visit to Facebook’s Prineville data center on Thursday, the temperature hit a high of 93 degrees outside
Building No. 1 of the data center was as noisy as an industrial factory, with air flowing through the cooling rooms and the humidifier room spraying purified water onto the incoming air at a rapid rate. It’s like peeking inside a brand new Porsche while it’s being driven at its fullest capacity.
Facebook’s data center here is one of the most energy efficient in the world. The social network invested $210 million for just the first phase of the data center, which GigaOM got a chance to check out during a two-hour tour.
Building No. 1 is where Facebook first started designing its ultra-efficient data centers and gear, and where it wanted the first round of servers that it open sourced under the Open Compute Project to live. Since then — Building No. 1 was opened up in the spring of 2011
Facebook has slightly tweaked its designs for Building No. 2 at the Prineville site, as well as the designs for its data centers in North Carolina and Sweden. Building No. 2 will use a new type of cooling system
Facebook is currently investing heavily in its infrastructure boom. It now has an estimated 180,000 servers for its 900 million plus users — that’s up from its estimated 30,000 in the winter of 2009, and 60,000 in the summer of 2010.
learning how to be an infrastructure company.
Open Compute Project Publishes Final Open Rack Spec
http://hardware.slashdot.org/story/12/09/19/2123219/open-compute-project-publishes-final-open-rack-spec
The Open Compute Project has published the final specification of the Open Rack Specification, which widens the traditional server rack to more than 23 inches. Specifically, the rack is 600 mm wide (versus the 482.6 mm of a 19-inch rack), with the chassis guidelines calling for a width of 537 mm. All told, that’s slightly wider than the 580 mm used by the Western Electric or ETSI rack.
Open Rack 1.0 Specification Available Now
http://opencompute.org/2012/09/18/open-rack-1-0-specification-available-now/
The Computer Science Behind Facebook’s 1 Billion Users
http://developers.slashdot.org/story/12/10/05/233254/the-computer-science-behind-facebooks-1-billion-users
“Much has been made about Facebook hitting 1 billion users. But Businessweek has the inside story detailing how the site actually copes with this many people and the software Facebook has invented that pushes the limits of computer science.”
Facebook: The Making of 1 Billion Users
http://www.businessweek.com/articles/2012-10-04/facebook-the-making-of-1-billion-users
Open Compute Project Hosts Hackathon
http://hackaday.com/2012/12/11/open-compute-project-hosts-hackathon/
The folks at Open Compute Project are running their annual summit in January, but this year they’ll be adding a hardware hackathon to the program. The hackathon’s goal is to build open source hardware that can be applied to data centers to increase efficiency and reduce costs.
Open Compute ‘Group Hug’ Board Allows Swappable CPUs In Servers
http://hardware.slashdot.org/story/13/01/16/2035207/open-compute-group-hug-board-allows-swappable-cpus-in-servers
“AMD, Intel, ARM: for years, their respective CPU architectures required separate sockets, separate motherboards, and in effect, separate servers. But no longer: Facebook and the Open Compute Summit have announced a common daughtercard specification that can link virtually any processor to the motherboard.
“Group Hug” Board Allows Swappable CPUs in Servers
http://slashdot.org/topic/datacenter/group-hug-board-allows-swappable-cpus-in-servers/
It’s finally happened: a swappable daughtercard will allow AMD, Intel, and even ARM CPUs to be interchanged, thanks to a new Open Compute specification.
AMD, Intel, ARM: for years, their respective CPU architectures required separate sockets, separate motherboards, and in effect, separate servers. But no longer: Facebook and the Open Compute Summit have announced a common daughtercard specification that can link virtually any processor to the motherboard.
AMD, Applied Micro, Intel, and Calxeda have already embraced the new board, dubbed “Group Hug.” Hardware designs based on the technology will reportedly appear at the show. The Group Hug card will be connected via a simple x8 PCI Express connector to the main motherboard.
It’s hard to overstate the potential of the technology. Although many other components within a server—including power supplies, hard drives, memory and I/O cards—have long been replaceable, processors have not.
while a standard has been provided, it may be some time before the real-world appearance of servers built on the technology.
Customization of servers will increase exponentially as a result
Intel, Facebook collaborate on new data center rack technologies
http://www.cablinginstall.com/articles/2013/january/intel-facebook-data-center-rack.html
Intel and Facebook announced that the companies are collaborating to define the next generation of data center rack technologies to enable the disaggregation of computing, network and storage resources. In a follow-on announcement, Quanta Computer unveiled a mechanical prototype of the companies’ new silicon photonics rack architecture, the better to demonstrate the potential for total cost, design and reliability improvements via disaggregation.
Rack disaggregation refers to the separation of those resources that currently exist in a rack, including computing, storage, networking, and power distribution into discrete modules.
The mechanical prototype from Quanta Computer is a demonstration of Intel’s photonic rack architecture for interconnecting the various resources, showing one of the ways computing, network and storage resources can be disaggregated within a rack. Further, Intel says it will contribute a design for enabling a photonic receptacle to the Open Compute Project (OCP), and will work with Facebook, Corning and others over time to standardize the design.
Who needs HP and Dell? Facebook now designs all its own servers
Facebook’s newest data center won’t have any OEM servers.
http://arstechnica.com/information-technology/2013/02/who-needs-hp-and-dell-facebook-now-designs-all-its-own-servers/
Nearly two years ago, Facebook unveiled what it called the Open Compute Project. The idea was to share designs for data center hardware like servers, storage, and racks so that companies could build their own equipment instead of relying on the narrow options provided by hardware vendors.
While anyone could benefit, Facebook led the way in deploying the custom-made hardware in its own data centers. The project has now advanced to the point where all new servers deployed by Facebook have been designed by Facebook itself or designed by others to Facebook’s demanding specifications. Custom gear today takes up more than half of the equipment in Facebook data centers. Next up, Facebook will open a 290,000-square-foot data center in Sweden stocked entirely with servers of its own design, a first for the company.
“It’s the first one where we’re going to have 100 percent Open Compute servers inside,” Frank Frankovsky, VP of hardware design and supply chain operations at Facebook, told Ars in a phone interview this week.
At Facebook’s scale, it’s cheaper to maintain its own data centers than to rely on cloud service providers, he noted. Moreover, it’s also cheaper for Facebook to avoid traditional server vendors.
Like Google, Facebook designs its own servers and has them built by ODMs (original design manufacturers) in Taiwan and China, rather than OEMs (original equipment manufacturers) like HP or Dell. By rolling its own, Facebook eliminates what Frankovsky calls “gratuitous differentiation,” hardware features that make servers unique but do not benefit Facebook.
It could be as simple as the plastic bezel on a server with a brand logo, because that extra bit of material forces the fans to work harder. Frankovsky said a study showed a standard 1U-sized OEM server “used 28 watts of fan power to pull air through the impedance caused by that plastic bezel,” whereas the equivalent Open Compute server used just three watts for that purpose.
A path for HP and Dell: Adapt to Open Compute
That doesn’t mean Facebook is swearing off HP and Dell forever. “Most of our new gear is built by ODMs like Quanta,” the company said in an e-mail response to one of our follow-up questions. “We do multi-source all our gear, and if an OEM can build to our standards and bring it in within 5 percent, then they are usually in those multi-source discussions.”
HP and Dell have begun making designs that conform to Open Compute specifications, and Facebook said it is testing one from HP to see if it can make the cut. The company confirmed, though, that its new data center in Sweden will not include any OEM servers when it opens.
Rackspace: Why we’re designing our own cloud servers
Just what will it take to compete with Amazon and Google
http://www.theregister.co.uk/2013/03/18/rackspace_server_fleet_open_compute/
Any cloud computing provider that wants to operate at scale and compete against its peers is under pressure to build some kind of custom hardware. It may, in fact, be necessary to compete at all.
That is what Rackspace, which is making the transition from website hosting to cloud systems, believes.
And that’s why the San Antonio, Texas-based company started up OpenStack – the open-source cloud controller software project – with NASA nearly three years ago, and accepted an invitation from Facebook to join the Open Compute Project, an effort by the social network to design open-source servers and storage and the data centres in which they run.
“We took the model of putting servers up on racks very quickly and turning them on in 24 hours and we called it managed hosting. At the time, all of our founders at Rackspace were Linux geeks and they were all do-it-yourselfers, and they were literally building white-box servers. They were buying motherboards, processors, and everything piecemeal, and we assembled these tower-chassis form-factors on metal bread racks and it was really not very sexy.”
“We mimicked what the enterprise would do in their data centre to go win business from those enterprises,” said Engates. “Enterprises didn’t want to think they were being put on a white-box, homemade server. They wanted a real server with redundant power supplies and all that fancy stuff.”
Rack servers evolved and matured
“Now,” said Engates, “we are basically back to our own designs because it really doesn’t make a lot of sense to put cloud customers on enterprise gear. Clouds are different animals – they are architected and built differently, customers have different expectations, and the competition is doing different things.”
What is good for Facebook is not perfect for Rackspace, as the latter explained at the Open Compute Summit back in January, but the basic rack and server designs can be tweaked to fit the needs of a managed hosting and public cloud provider.
Rackspace has been pretty quiet about what it has been doing with Open Compute up until earlier this year, and part of that is Rackspace’s decision to radically change its business with both OpenStack and Open Compute.
The custom Open Compute machines cost anywhere from 18 to 22 per cent less to build than the bespoke boxes from HP and Dell that make up about 18 per cent of the Rackspace server fleet, which is about 16,300 of the 90,525 boxes that were running at the end of December across the cloud company’s data centres.
“We have all kinds of horses in the barn, but for the past 18 months, we have been only dealing in high-density compute with boxes completely maxed out. We will run almost 20 kilowatts per cabinet, so it is very dense compute,” said Roenigk.
Facebook revealed as company behind $1.5 billion Altoona project
http://www.desmoinesregister.com/article/20130419/BUSINESS/130419046/Facebook-revealed-company-behind-1-5-billion-Altoona-project?gcheck=1&nclick_check=1
Facebook is the company behind a $1.5 billion data center that’s expected to land Altoona, a project that’s being touted as “the most technologically advanced data center in the world,” legislative sources told the Des Moines Register today.
The project is expected to be completed in two $500 million phases near Altoona, where leaders have provided a green light for a 1.4 million square foot facility. When completely built out, experts expect the facility will cost $1.5 billion.
The Iowa Economic Development Authority Board and Altoona’s City Council are expected to consider incentives for the project on Tuesday.
Penguin Computing to make Open Compute servers
And apparently a lot more money, thanks to Zuck & Co.
http://www.theregister.co.uk/2013/05/10/penguin_computing_open_compute_servers/
Linux server and cluster maker Penguin Computing is a member of the Open Compute Project started by Facebook to create open source data center gear, and now it is an official “solution provider”.
This means that Penguin now has OCP’s official blessing to make and sell integrated systems based on the motherboard and system designs that other OCP members cook up, and that it has met the manufacturing criteria to be able to put the OCP label on the machines.
Not everyone can use the stock Facebook designs, which come with their own custom Open Rack and power distribution and battery units, and which are tied very tightly to the custom data centers that Facebook has built. Some companies cannot easily ditch 19-inch racks and want some of the benefits of the “vanity free” OCP designs without having to gut their data center or build a new one. And that’s where companies like Penguin will come in, helping tweak OCP designs to fit into existing data centers.
“The world is changing, and coupled with some of the work we have done with ARM-based servers, we just want to be at the front end of these changes,” says Wuischpard – and the official designation by OCP means that Penguin can sell machines into Facebook itself.
“Our estimation is that OCP could be 40 per cent of our business within the next twelve months,” Wuischpard says, “if not more.”
That assumes that HP and Dell don’t get all gung-ho about Open Compute machinery. And if they see Penguin making money, they just might have no choice but to embrace open hardware, no matter how much they want to sell their own designs.