As the need for fast, large-scale computing to power sites like Facebook or even computing clouds has grown, manufacturers such as Rackable Systems are taking notes on server design from Google, which builds its own systems. The goal of their mimicry is to provide more computing power in a smaller form factor while using less energy.

An article in EEtimes today details the emergence of these Google-inspired servers, which include features such as heat-tolerant processors to save on cooling costs, a focus on motherboards containing 12v-only power supplies for servers, putting two servers on one board and stripping out unnecessary parts.

These are all ways Google apparently modifies its boxes to deliver information faster and more cheaply. Rackable’s new CloudRack servers will offer dual servers on one board that crams more computing power into a smaller space, as well as 12V-only motherboards. The use of only 12 volts on a motherboard is supposed to make the power supplies more efficient by reducing the energy lost when having to convert electrical current to run at various different voltage levels.

IBM’s iDataPlex servers, designed for the cloud, have stripped away unnecessary hardware — a move aimed at reducing power-consuming components and saving space. Heat-tolerant processors allow a data center operator to keep air conditioning bills down, saving as much as 4 percent of total energy costs for each degree dropped. So as computing requires more scale, Google’s innovations influence other buyers and sellers of technology even as the search giant slows its own data center construction.

Yahoo, one of the largest websites on the planet is being plagued by series of problems related to Domain Name System (DNS). A test using Gomez’s testing service shows error messages in certain cities such as Chicago. Others are experiencing slower access to Yahoo websites including the home page. If you have experienced these problems do let me know in the comments. I am going to try and talk to Yahoo and get to the bottom of it as quickly as I can.

Today Amazon Web Services launched the beta version of its content delivery network service called CloudFront. As Om mentioned in September when the service was announced, this is a good move for Amazon, and something that may put the hurt on fellow CDNs such as Limelight and Akamai. Amazon will charge a usage based fee, rather than a long-term contract, bringing CDN prices even lower for smaller web players who don’t have the scale to negotiate lower prices. Here’s how it works from the release:

The service caches copies of content close to end users for low latency delivery, while also providing fast, sustained data transfer rates needed to deliver popular objects to end users at scale. CloudFront works seamlessly with Amazon S3, where users store the original versions of objects delivered through the service. Customers need only put their objects into an Amazon S3 bucket and then register that bucket with the new service using a simple API call, which then returns a domain name used to access content through the network of edge locations.

With prices ranging from 17 cents per gigabyte for the first 10 terabytes sent out a month, to 9 cents per GB for everything over 150 TB, the service seems to undercut the pricing offered by other CDNs for small to medium sized customers. It might be a good thing that Akamai’s looking at advertising.

It was a matter of when, not if, Amazon would launch a content delivery business in addition to its current suite of web services that include S3 storage service and EC2 on-demand computing. The Seattle-based company has announced its intention to offer a content delivery service that could shake the very nature of the industry and pose a serious challenge to not only dozens of CDN upstarts but also become a thorn in the side of existing giants such as Akamai Technologies and Limelight Networks.

In an email to its customers today, Amazon said that the service will be available later this year and will utilize the company’s points of presence in North America, Europe and Asia.

This new service will provide you a high performance method of distributing content to end users, giving your customers low latency and high data transfer rates when they access your objects. The initial release will help developers and businesses who need to deliver popular, publicly readable content over HTTP connections.

Ironically, Amazon was beaten to the CDN punch by New York-based Voxel that started offering CDN services based on Amazon’s S3 service. ”We are announcing this right now because we want to give a heads up to our customers,” said Adam Selipsky, Vice President of Product Management and Developer Relations for AWS. It is more like putting their competition on notice, but Adam was too polite to say that. “It is a more horizontal and broad offering.” In other words, while it is not going to replace Akamai tomorrow, it is going to make CDNs affordable even for the tiniest startup, without major cash outlays. 

Why is this service disruptive? Amazon is going to bring a level of transparency to a business that has a sales model much like an brokerage firm in the 1980s. Amazon wants to make buying CDN services as simple as buying a book. Amazon executives told me that company is going to be charging its customers on usage instead of long-term contracts current players foist on their clients.

In addition, the company will publish its prices on the web and most importantly it is going to be inexpensive.And that will make the service even more attractive to hundreds of small companies who are already using Amazon Web Services for their web operations, who don’t want to sign long contracts with CDN operators. When I asked Tal Saraf, General Manager of the AWS Content Delivery Service, if the company expected the video-delivery to be one of the most used service, he said the company expected to delivery all sorts of content including web-objects (images, JavaScripts etc.)

You’ll start by storing the original version of your objects in Amazon S3, making sure they are publicly readable. Then, you’ll make a simple API call to register your bucket with the new content delivery service. This API call will return a new domain name for you to include in your web pages or application. When clients request an object using this domain name, they will be automatically routed to the nearest edge location for high performance delivery of your content. It’s that simple.

Amazon executives declined to talk about the pricing. “We will talk about the pricing when we launch the service,” Selipsky said. He declined to comment on the impact of their pricing will have on their competitors – nearly two dozen content delivery networks – and how much their business is going to suffer. DowJones Venture Source estimates that from 2005 through second quarter of 2008 nearly $980 million was invested in content delivery companies.

If Amazon delivers what it is promising – a simple API based CDN – then it would put then not only ahead of all CDN players, but also force rivals to meet the rules (and pricing) set by Amazon. There is a good chance that it is going to drive weaker players right out of the game. 

My final take on this news: Akamai is less likely to be impacted in the near term, but it further commoditizes the CDN business and forces a big shakeout in the industry, taking down the small and the weak. Akamai has been focusing on value add services, as a way to stay ahead of the commoditization of the basic CDN services.

Earlier this month, the blogosphere was lit up with stories of Google filing a patent around the concept of data centers on the sea that use wave power — essentially retrofitted ships and barges that would be docked 3-7 miles from shore, in 50-70 meters of water. While they might sound like something an evil villain in a James Bond thriller would build, the concept is actually not that far-fetched. In fact, one stealthy startup is currently working on building one of its many data centers on the sea: International Data Security.

All told, San Francisco-based IDS plans to build 50 such floating data centers — 22 in the U.S. and the rest worldwide. Old, decommissioned ships will be retrofitted and permanently housed at a pier where they will be connected to the Internet via fiber and get power from traditional utility connections. The company is also going to have microwave wireless backup connections.

IDS was started by Ken Choi, who is also the CEO, while Richard Noughton, a former U.S. Navy admiral, is the president. From what I’ve been able to learn, the company’s first ship will be docked on Pier 50 in San Francisco. And because the data center ship will be on the water, the theory goes, less power would be needed to cool the entire facility, which would in turn lower the costs of operating the data center.

Richard Donaldson, COO of UnitedLayer, a San Francisco-based managed hosting provider, turned me on to IDS. The news of its floating data center ambitions was first reported by our friend Rich Miller. Since then I have pieced together some more information.

There are many caveats around IDS and its plans. For instance, it is not clear how much funding the company has available, and when it will launch the proposed data center in San Francisco.

Some critics believe that data centers in salt water, surrounded by salty air aren’t such a good idea. Power is another issue: While a ship can float away to a safer location in case of a problem, the ship would need considerable backup power to keep the servers on. Even then, it couldn’t really go too far away from the microwave towers that feed backup bandwidth.

Kooky as these ideas of floating data centers seem, data-center industry executives think that they are a good way to deal with the real-estate crunch in markets where demand for data centers is at a premium. San Jose, San Francisco and New York are three such examples, mostly because large corporations and web companies want to house their machines in buildings to which they can drive.

Anyway, stay tuned as I try and dig up more on the progress of IDS and how close they might actually get to their floatilla of data-center ships.

It is not clear how old Google is - some argue that world’s largest search engine operator is 13 - after all it operated in stealth for about 3 years before launching in September 1998. Many major news organizations are going with September 2008 as the tenth anniversary so I am going to play along. Forbes.com even asked the question, Has Google Changed The World? from many well known people. For some odd reason they decided to seek my thoughts.

Gandhi changed the world. The steam engine changed the world. Heart transplants changed the world. The Internet changed the world. Google simply made a small (albeit important) contribution toward making Internet a better experience for all of us.

Google’s contributions are still worthy of praise. It is no longer impossible to find relevant information on the fast-growing Internet. I remember tearing my hair out looking for relevant information. Today it is as simple as acting on our impulse to seek that knowledge–and that has infinitely changed the way we interact with the machines.

The article triggered a chain reaction and a trip down the memory lane. I had been a Google-addict for a while and loved its simple elegance over rivals such as AltaVista and Inktomi-powered searches. I had talked to the company earlier, but I didn’t meet the Stanford duo in person up until September 1999. The company had just raised about $25 million in venture money.

“I have never paid more money for so little a stake in a startup,” John Doerr of Kleiner Perkins Caufield & Byers was heard saying. Good thing he did - for he paid next to nothing for what could arguably be the Internet-equivalent of Alaskan oil and gas fields.

Larry Page & Sergey Brin had stopped by at the Forbes.com offices and we talked at length about the company. It ultimately resulted in this feature, How Google Is That? Larry still had the same disastrous haircut he supports today. Brin was measured and logical as always in his responses. They thankfully made no meaningless and “do-no-evil” hypocritical statements. They were just two guys out to change the world. I remember getting along with them famously, but never saw or talked to them since, though I have been to many Google press events.

Then & Now: You’ve come a long way baby

The company was 12-months old. They had just come up with their version of contextual-text advertisng system. They had 40 employees, were looking for an inhouse chef, and were doing about 4 million page views a day and about 4 million searches a day. That’s 45 searches per second. No one in the company owned a glider, though their venture backers had their own private planes. The company was housed in 165 University Avenue in Palo Alto and the co-founders were single.

In July 2008, Google registered 7.23 billion searches - about 242 million a day. That works out to about 4 10 million searches in an hour or over 1100 2772 searches per second. (Funny, it turned out to be much bigger than the market estimates used by Google.) It had sales of $5.4 billion in the second quarter of 2008 alone. It now employs over 19,000 people. Larry and Sergey are billionaires and own a Boeing 767 & a Boeing 757. They are both married. The company has offices in multiple locations and data centers that are sprinkled around the globe.

After meeting with them and discussing the merits of search-only approach versus portals, I came to this conclusion: “Perhaps the other Stanford duo, Yahoo! cofounders David Filo and Jerry Yang, should be a little concerned–their media ambitions have superseded their customers’ desire for a really smart search engine.” In hindsight, I am surprised I was able to get away with making that statement and my editor didn’t catch what clearly was an opinion - a no-no in the non-blog mediascape. After all, it seemed so stupid to suggest that because Yahoo had 240 million page views a day and was literally printing money.

Brin tried to convince me that the text-based contextual advertising (first popularized by LinkExchange, a company that was bought by Microsoft) was their way of making money. “Banners are not working and clickthrough rates are falling, I think highly focused ads are the answer,” Brin said, and pointed out that Google would be in black in 24 months. By 2001, I could have kicked myself for doubting the kid!

Why did they win?

Fast forward 9 years, and most of Google’s competitors have gone to the great technology graveyard, nary a tombstone. Simpli.com, Dogpile, Direct Hit and Northern Light were all part of the new search engines that were taking on the incumbents like Yahoo, Lycos and HotBot and wanted to make web searches simpler and more accurate.

“Google is essentially trying to categorize and catalog the web. We have a very different product and a different approach,” Jeffrey Stibel, cofounder and CEO of then Providence, R.I.-based Simpli.com told me for the Forbes.com story. He was taking a more exotic linguistic approach to search. It is now owned by Valueclick, an ad-network.

In comparison, Google’s analysis of the link structure of the World Wide Web and large-scale data mining and ability to ranks a page against similar pages turned out to be the right approach. Was it just the algorithm and a better monetization scheme? Was it a right solution at the right time? I think it was a bit of all that - but most importantly, it was a farsighted approach to infrastructure and the network.

It’s the infrastructure stupid.

This was the critical difference - I wrote about it recently - between winning and losing. I was reminded of this by an old PowerPoint presentation. They talked about using commodity compute infrastructure to out muscle everyone and doing analysis of the web like it has never been done before. It seems so obvious today - but back then it was an idea ahead of its time. The impact of pizza box servers was yet to be seen, and companies like Cobalt Networks (sold to Sun Microsystems for $1 2.4 billion) were selling early versions of Linux-powered thin servers, but they were not cheap by any means.

Many on Wall Street question why Google spends so much money on infrastructure. The question is why not - after all every millisecond of performance means more searches and more searches mean more advertising. More infrastructure means more crawling, more indexing and better results. I think that slide reminds us of the fact that infrastructure-as-an-advantage is in the DNA of Google. And that is unlikely to change - and that is why world’s smartest engineers and computer scientists still want to work there.

History has made a genius out of all who bet on Larry and Sergey - the investors, the employees, journalists who were enthralled by their story. In reality to those who built Google, it was the only option.

Tomorrow: What You, Me & Corporations Can Learn From Google

It’s almost becoming routine, these outages at Google’s Gmail service. After we reported last week that there were some problems with Google Apps, today a much bigger outage hit Google’s email service, taking down the entire system, it seems.

Given that our company relies on Google’s Gmail and GTalk service, our operations came to a standstill this afternoon. We aren’t a large company but the losses are very real, especially in productivity. I wonder how the big customers of Google — folks like Sanmina-SC — are dealing with this e-blackout.

If an outage of this magnitude can strike Google, the company with a fearsome infrastructure, I wonder who — if any — can plan for the worst. I guess it’s time to stop picking on Twitter, which was fast becoming synonymous with the word “outage.” In fact, in recent weeks, not only Google’s services but those of Amazon’s S3 and Apple’s MobileMe have gone on the blink, leading us to rethink our assumptions about the reliability of the web as a platform. Clearly a lot of work still needs to be done.

How is your workday being impacted by Gmail outage? Share your stories with us.

Photo courtesy of Kyle May via Flickr.

In Beijing, Internet access will soon be in high demand: Half a million people are expected to visit the city of 17 million for the Olympics, and most of them will want web-based access to personal and corporate sites. This may well be the largest international remote access event ever. Much of the attention has been around whether visitors can surf the Internet. But some people are wondering whether they should. Is it safe to surf from China?

“With Software-as-a-Service applications, more users will access their applications across the Internet, so companies can’t rely on physical or firewall access,” said Marc Gaffan, director of product marketing for RSA’s Identity and Access Assurance Group. “The risks are significantly increased.” The U.S. government’s head of counter-espionage, Joel Brenner, is also cautioning travelers to Beijing about identity theft and other threats.

Most users assume that a secure web connection makes them safe. After all, that little yellow SSL padlock doesn’t just mean your traffic is encrypted, it also tells you the URL you’re visiting is the one you wanted — right? Not always, said Jayson Agagnier, a security consultant who specializes in corporate counter-espionage. “On older browsers, the padlock will still be there even if the user accepts a certificate that is not publicly signed.”

To collect passwords, hackers only need to trick surfers into logging in. Many casual users won’t think twice about typing in www.mybank.com and being redirected to mybank.login.com, provided that the new site looks the same. “Obtaining a certificate is fairly easy,” said Gaffan, “and no one really checks the certificate in the lock.”

Phishing for usernames can happen anywhere, but when half a million people descend upon a country that heavily regulates its Internet, it’s an excellent opportunity for mischief. So how can organizations protect themselves? Here are some suggestions:

• Have vacationing workers check URLs closely to be sure the site they’re on matches what they entered, even if it looks the same.
• Get a more trusted — and more costly — Extended Validation certificate. These are harder for a fly-by-night operation to get because they require more thorough background checks.
• Use dynamic passwords that change every minute, so even if someone intercepts a password it quickly expires.
• Use “fat client” VPNs based on IPSEC or SSL instead of relying on a secure web login. VPN clients can’t be tricked into thinking they’re at the right site.

Capturing logins isn’t the only risk, however. It would take a real conspiracy to present a completely faked site, complete with the right URL and a valid SSL certificate. But if a government owns the network, it’s the lawful man in the middle, and it has the resources for such schemes. “You can control the DNS, display any page you like, entice people to log in,” said Gaffan. As IOC president Jacques Rogge said on July 31, “We are not running the Internet in China. The Chinese authorities are running the Internet.”

Agagnier says Olympics-related travel presents a huge industrial and economic espionage opportunity, but Gaffan says he thinks an elaborate network attack may be more work than it’s worth. “If I were a fraudster, I would just spend two hours in Beijing hotels and Internet cafes installing key loggers. You could collect names and passwords, even things like frequent flier numbers that could be used for corporate espionage to track the travel patterns of a competitor’s employees.

Syntenic CTO Daniel Koffler agrees: “I would be concerned about malicious WiFi access points … You don’t really need to own the back-end pipe; a cheap access point and an SSL proxy is all anyone on the street would need to collect some serious information. While you’re in Beijing, if the state wants your data, they’re going to get it. It’s the billion or so citizens you have to watch out for.”

Perhaps the best defense is to take the week off. Several enterprise IT professionals I interviewed for this story said they’re simply telling their users not to log in from China.

There’s already a ton of activity taking place in the cloud computing space, so much so that it can be hard to know who to watch. In many cases, it’s too early to pick winners. But there are distinct sectors of the IT industry that are particularly well suited to the on-demand, pay-as-you-go economics of cloud computing.

Here are eight segments — and one company that’s a segment all its own — that we’re tracking closely.

Hosting companies that make the jump: When it comes to reliable managed hosting, Rackspace leads the pack. (Its VMware-based Mosso offering may appeal more to enterprises trying the cloud for the first time.) Clouds like XCalibre’s Flexiscale and Joyent are already there, but don’t have Rackspace’s installed base.

Stack-specific clouds: While Google and Amazon get the headlines, Engine Yard is heavily involved in the Ruby on Rails development community. Competitor Heroku is also Rails-focused, but relies on Amazon for its hosting platform.

Tools to wrangle virtual machines: To manage your EC2 machines, you’re going to need help. RightScale makes software for managing machines in the cloud; its tight focus on Amazon has made it an early favorite. Elastra, Enomalism and others have similar solutions.

Testing sandboxes: For many enterprises, a testing sandbox is the perfect way to start using on-demand infrastructure. CohesiveFT’s Skytap (a sister to Flexiscale) spins up testing machines in a cloud, but incumbent Surgient and recent entrant StackSafe aren’t far behind. And once you’ve tested a machine and seen that it works, why not leave it in the cloud?

Cloud-based development platforms: Companies like Rollbase and Coghead let non-developers build data-driven applications of any sort (as opposed to more specialized platforms like those of Salesforce and Ning.) But Intuit’s Quickbase, which now has access to Quickbooks data, has a head start: Millions of small businesses. Is this how SMB gets cloud?

Scaling frameworks: Wall Street needed fast, reliable applications that grew easily. Instead of adding more, bigger servers, they used Gigaspaces to bundle whole server clusters into discrete “processing units” that can be cloned to add capacity. In addition to being faster and scaling better, these units don’t care whether they’re in a private data center or a cloud.

Application delivery networks: What has tens of thousands of servers worldwide, a global network connecting them, and isn’t Google? Akamai. What was once a way of getting bits to far-flung corners of the Net is an often-overlooked cloud: Akamai has been able to run code at the edge since 2000. Its 2007 acquisition of Netli made it matter to enterprises even more. Akamai can weather heavy load and may be able to withstand attacks better than centralized clouds.

Cloud builders: 3Tera lets companies get into the cloud business. Enterprises can make in-house clouds on existing data centers; or service providerscan build their own cloud offeringsin the way Enki and others have. In 3Tera’s model, subscribers drag and drop the firewalls, servers and appliances they need. The company’s software then maps these virtual application stacks to servers and network segments. The results are impressive: On seeing 3Tera for the first time, ESM guru John Willis was so impressed he insisted on logging in to the icons on his screen to verify that it wasn’t just a demo.

The obvious one: Of the three big virtualization firms, only one (Microsoft) also has millions of desktops, two handset platforms, licensing for desktops, servers and applications, synchronization, and a huge online presence. Up until now, the Redmond giant has been treading carefully; it has to convert billions of dollars of shrink-wrap sales to on-demand revenue streams. But Microsoft’s going to be a huge player in the cloud.

For more insights into cloud computing trends, check out the recent GigaOM/Bitcurrent briefing on cloud computing that was launched at Structure 08.

When we recently heard about the history of YouTube’s growth strategy from CEO Chad Hurley’s point of view, he described it as “hanging onto a rocket.” But an engineer’s take is always going to be a bit less rose-colored and a bit more about the terrifying situations you brained your way out of. So we were particularly interested to tune in to a talk at YouTube’s developer conference Thursday by Cuong Do, an early software engineer who’s now manager of the site’s Core Product Engineering group.

Do’s talk was titled “Behind the Scenes: A Look Into YouTube’s Infrastructure,” and he didn’t disappoint, with harrowing tales of outages; gory details about the specific languages, architectures, and tools YouTube uses; and a flow-chart level view on the way the site handles uploads and video delivery while undergoing the massive usage it sees on a daily basis.

“One of the key phrases we had in the early days was ‘These are good problems to have,’” Do said. “And after a while we’re like, ‘I’m going to kill the next person who says that.’”

YouTube promised it would post video from the talk on its site eventually, but I don’t see it there yet, so check out the version from my handheld camera.