Facebook is famously green on the data center side, and today, it’s introducing a new level of transparency. With two new dashboards, you’ll now be able to see the real-time impact and efficiency behind the scenes of all your Facebook activity.

Every like and comment and photo shared on Facebook has some ecological cost. The machines that process and store them use power, which still mostly comes from coal; and they need to be cooled. Facebook and the Open Compute coalition of hardware manufacturers and tech companies have made great strides in designing the most efficient machines possible for doing this job with as little negative environmental impact as possible.

In fact, Facebook has two data centers dedicated to Open Compute’s green-first designs. One is in the cooler clime of Prineville, Ore.; the other is in Forest City, NC.

Both data centers now have real-time online dashboards anyone can access (check out the Prineville dashboard and/or the Forest City dashboard for yourself — both are interesting and interactive).

On these dashboards, you can look at a visual representation of important data-center stats — the humidity, the power usage effectiveness (PUE), the water usage efficiency (WUE), and the temperature. You can also track these metrics over time throughout the past 24 hours, the past week, the past quarter, or even the past full year.

“We’re proud of our data center efficiency, and we think it’s important to demystify data centers and share more about what our operations really look like,” writes Facebook efficiency project manager Lyrica McTiernan in an Open Compute blog post today.

“We began sharing PUE for our Prineville data center at the end of Q2 2011 and released our first Prineville WUE in the summer of 2012. Now we’re pulling back the curtain to share some of the same information that our data center technicians view every day.”

For even more visual stimulation, here’s a collection of photos we took during a walking tour of Facebook’s Prineville data center. Especially interesting: How Facebook manages to keep the machines running at top efficiency without traditional air conditioning.

---

Big Data and Predictive/Real-time Analytics startups: Are you looking to jumpstart development & accelerate market traction? Sign up for the SAP Startup Focus program to receive technology, support, resources and community to help you develop new applications on SAP HANA, a cutting edge database platform. Get started here, and enter promo code “VB2013″ on the form.

---

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

Fusion-io, a leader in solid-state storage devices for high-volume datacenters, has acquired ID7, a maker of a Linux-based software-defined storage subsystem.

ID7′s signature product is an open-source project called SCST, the Generic SCSI Target Subsystem for Linux (we’re not quite sure how that results in the acronym SCST, but that’s how they do it). Fusion-io has already renamed ID7 as Fusion-io ID7, and affirmed its ongoing commitment to open source.

“This endeavor matches an overwhelming trend in the industry towards open industry standard server systems, whereby customers can choose the hardware platforms that best suit their needs and couple them with the most capable software to maximize their use,” Fusion-io’s blog post states. A bit further on, it adds, “Fusion-io enthusiastically supports open source, and we will continue to encourage the open principles that have made ID7 and SCST successful, including maintaining an open source version of SCST.”

ID7 chief technical officer and co-founder Mark Klarzynski published his own blog post today as well. It’s sort of a mash note to Fusion-io and its “laser-point focus aimed straight at the user.”

In addition, the move toward software-defined storage is an industry trend, with enterprise vendors including Nexenta, Scality, and SwiftStack attracting investments to pursue product development in this sphere. It’s symptomatic of a larger shift away from big, inflexible storage area networks (and data center hardware more broadly) and towards more flexible, software-defined computing resources that companies and data center providers can allocate as needed, in real time, without worrying about their physical locations, specific wiring, or underlying hardware.

The companies did not disclose the terms of the deal.

Via ZDNet

Photo: Random Fusion-io product shot. 

Microsoft and Symantec researchers busted into two data centers in New Jersey and Virginia today to shut down servers associated with a botnet called Bamital.

The companies had an order from the U.S. District Court in Alexandria, Va., according to Reuters, allowing them to enter the data centers. Those at the New Jersey facility seized one of the servers and shut it down. Others in Virginia convinced workers to contact their Netherlands-based parent company and shut down a server there.

The botnet secretly used victim’s computers to steal advertising revenue. The victims didn’t know this was going on until their computers were suddenly unable to search the Internet. Microsoft and Symantec say those people were served with a message that read, “You have reached this website because you computer is very likely to be infected by malware that redirects the results of your search queries. You will receive this notification until you remove the malware from your computer.” It then offered ways to do so.

Bamital is believed to have infected somewhere between 600,000 and 900,000 computers. Microsoft and Symantec are confident they’ve shut down the entire botnet, but notes that only “time will tell,” according to Microsoft’s Digital Crimes Unit’s associate general counsel Richard Boscovich who spoke with Reuters.

This isn’t Microsoft’s first go-around at taking down a botnet, however. In 2011, the company took down the Kelihos botnet, which is small by comparison at on 41,000 infected computers. At the time Microsoft said, however, that the botnet was capable of sending out over 3.8 billion spam emails a day. Microsoft also named suspected perpetrators behind the botnet including a man named Dominique Alexander Piatti.

Later, in 2012, a second and bigger Kelihos botnet was found in the wild. It was subsequently shut down by Russian security firm Kaspersky Lab.

Batman image via kevin dooley/Flickr

SANTA CLARA, Calif. — This morning, Facebook and a slew of big names in cloud computing and data center hardware unleashed a whole boatload of news — new hardware designs, new jobs, and new partnerships — all around the Open Compute Project.

AMD and Intel both showed off some new hardware products. Rackspace announced it has customized its own server hardware around OCP designs — and contributed its customizations back to the project. And Fusion-io was talking about its new 3.2TB ioScale card, also an OCP product.

The Open Compute Project is the Internet and hardware industries’ attempt to make computing vastly more efficient by pooling knowledge and resources. Close competitors are actually working collaboratively to come up with a better motherboard, a better power supply. And it’s not just industry titans, either; anyone can download and modify the specs for Open Compute hardware.

Why should anyone with a day job and a social life care about this?

Two reasons: First, every time you click anything anywhere on the Internet, it takes a tiny toll on the environment — something with the impact of an ant’s footstep on a boulder. It takes electricity to power the machines that process the clicks and carry the data around the world, and most of that energy ain’t coming from wind farms.

Second, every click, every bit of data costs the companies behind web services a tiny amount of money, maybe hundredths of a cent, maybe thousandths of a cent. They’re paying to power the services, to store the data, and to employ the folks who keep the machines running.

In aggregate, those tiny costs add up to tons and tons of carbon and billions of dollars in bottom-line costs for companies like Facebook, Amazon, et cetera. So by putting aside their differences and competitiveness just long enough to make better, faster, cheaper, kinder-for-the-planet servers, these companies are ensuring a better environment for all of us as well as lower operating costs (and higher profits) for themselves — something Facebook in particular can’t afford to overlook.

This open-source hardware movement is still in its early days. Today marks the start of the Project’s fourth Open Compute Summit; while the first Open Compute efforts were driven by Facebook, today’s summit is thick with IT corporations from around the world.

“We’re seeing signs that the industry is changing, it’s becoming more open,” said Facebook hardware chief Frank Frankovsky (pictured above and below) today in his opening keynote. “Suppliers are seeing that there’s a radical change in this space.”

And as hardware becomes more open, not unlike the world of software, where open-source is the backbone of most systems, Frankovsky and others try to look into the near future and figure out where Open Compute and its network of partners should go next.

“Where it’s all headed in my mind is about breaking up the monolith … disaggregating the system design,” he said.

What that means, he continued, is changing the way data center hardware works. Instead of moving slowly and changing out whole stacks of hardware, Frankovsky says hardware should become — and is becoming — more flexible to meet consumers’ needs, more customizable to different configurations, and more quick to adapt to innovation.

“Today, one of the biggest challenges in hardware design is trying to predict where the software’s gonna be,” said Frankovsky. “There’s an impedance mismatch between the speed at which software moves and the speed at which hardware can move.”

He said “smarter technology refreshes and upgrades” is another challenge. So for OCP, breaking up the monolith means malleable configurations for data center gear, smarter tech upgrades, faster innovation for speedier components, and above all, openness.

New OCP members announced today include players in storage, telecom, and microprocessors: SanDisk, EMC2, Fusion-io, HGST, ARM, Tilera, Calxeda, NTT Data, and Orange. Cole Crawford, a Linux Foundation advisor and former Nebula exec, has been named OCP’s COO and is its first full-time employee.

Another exciting news item is the opening of OCP’s first international chapter, OCP Asia Pacific. With partners like Tencent, Baidu, and Alibaba, OCP already has a significant foothold in Asian markets, which are themselves hugely important hubs for hardware design and manufacturing. Interest in the region has been so strong that Frankovsky said some interested parties had taken it upon themselves to translate OCP specs into Japanese. Plus, he pointed out, “Asia’s going to have their own way to focus on their own issues,” such as earthquake tolerance and high-density physical spaces.

For Facebook’s part, the company is open-sourcing cold-storage versions of OpenVault and OpenStack, which are in use in its new-ish Sub-Zero cold storage facility, as well as Dragonstone, a new database server for its Swedish data center featuring high availability with dual motherboards and power supplies.

For more specifics, check out the OCP’s blog post on today’s announcements, and stay tuned for more news from the Summit.

---

Big Data and Predictive/Real-time Analytics startups: Are you looking to jumpstart development & accelerate market traction? Sign up for the SAP Startup Focus program to receive technology, support, resources and community to help you develop new applications on SAP HANA, a cutting edge database platform. Get started here, and enter promo code “VB2013″ on the form.

---

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

Flash storage business Fusion-io has launched its new ioScale product line, which promises to make all-flash-memory data centers a more feasible proposition for companies.

Ideally, ioScale makes the all-flash-drive data center more practical for companies that rely on high-performance hardware. Each one of ioScale’s units provides up 3.2 terabytes of Fusion ioMemory capacity, with prices starting at $3.89 per gigabyte. By offering ioScale to companies who order a minimum of just one hundred units, Fusion-io likely hopes to outgun competitors like SolidFire.

Fusion-io claims ioScale’s big benefits include:

• Up to 3.2 terabytes of capacity on a single half-length PCIe slot, which can enbable scaling to 12.8 terabytes or more and reduce the need for disk drive bays
• Hyperscale servers supporting UEFI
• High level endurance in all capacities
• Compatibility with the Fusion ioMemory SDK

Fusion-io CEO David Flynn wants the ioScale product line to be there for forward-thinking and young businesses that are willing to embrace an all-flash-memory approach for data centers, specifically “hyperscale and cloud companies.”

“By making ioScale available to growing webscale and emerging cloud companies, Fusion-io is at the forefront of the transition to the all-flash hyperscale data center, powered by open-software-defined solutions,” Flynn said in a statement. “Hyperscale companies are an entirely different market with different needs compared to enterprise organizations.”

Salt Lake City-based Fusion-io was founded in 2006.

Data center photo via Supri Suharjoto/Shutterstock

Google is telling Asian users to expect Google services to run up to 30 percent faster once a trio of new data centers come online.

Google’s head of products in India, Lalitesh Katragadda, told India’s Economic Times that it expects data centers in Singapore and Taiwan to be operational in 2013. A third data center in Hong Kong is also under construction.

As a result, Google services users in most parts of Asian, including the Indian subcontinent, can expect faster, more responsive performance from their Google-made web apps.

While India is one of Google’s largest markets, Katragadda said the climate wasn’t suitable for data centers (odd, considering Google competitor Facebook has had such marked success in hotter weather and higher temperatures). Still, the company hopes to boost slow speeds for services like YouTube and Gmail with relatively nearby facilities.

“Internet connectivity speed in India is not very high. These data centers will be crucial to this market due to its proximity,” said Katragadda.

“More new Internet users are coming online everyday here in Asia than anywhere else in the world,” said Google of its new facility in Changhua County, Taiwan. The entire project, which it expects to complete in the second half of 2013, will cost around $300 million by the time it’s finished.

“This data center will be the first in our fleet to save energy through a nighttime cooling and thermal energy storage system,” reads the project’s page. “And, like our other facilities in Asia, this will be one of the most efficient and environmentally friendly data centers in the region, built to the same high standard we use around the world.”

The Hong Kong data center is still in formative stages, just more than a year after Google bought the 2.7 hectares of land in the Tseung Kwan O Industrial Estate in Kowloon on which it’s constructing the new facility. The Singapore facility will be the first of the three to start operations, flipping its switches in the next couple months.

Interestingly, Google is launching community grants in each area where the new data centers will be opened.

Generally, Google claims its data centers use 50 percent less energy than those of its competitors, which for various products include Facebook, Amazon, Yahoo, Microsoft, and dozens of other companies.

top image via Google

---

Big Data and Predictive/Real-time Analytics startups: Are you looking to jumpstart development & accelerate market traction? Sign up for the SAP Startup Focus program to receive technology, support, resources and community to help you develop new applications on SAP HANA, a cutting edge database platform. Get started here, and enter promo code “VB2013″ on the form.

---

PN claims to be driving the “third wave of virtualization into the ethernet hardware market.” IT administrators can use the systems to achieve high-performance computing and bolster private and public cloud data centers.

Menlo Ventures led this round of $23 million.

“Pluribus Networks’ Server-Switch architecture is a revolutionary rethinking of the infrastructure required for today’s new datacenters,” said Mark Siegel, the managing director at Menlo Ventures, in a statement. The need for a virtualized and programmable network fabric is imperative to dynamically allocate and manage connectivity and data flows.”

This is not Menlo Ventures’ first investment in a virtualization startup. It invested in Tintri over the summer, a company that produces Flash storage appliances for virtual machines. Previous investors New Enterprise Associates and Mohr Davidow Ventures also joined this round, following a $17.5 million Series B led by NEA in April 2011.

Now with more than $40 million in its stocking, Pluribus Networks will have a very Merry Christmas and a strongly supported start to the new year. Read the press release.

IBM is announcing today it has created breakthrough silicon nanophotonics chips, or those that can use traditional silicon technology and laser light to  transfer data at extremely high speeds. These components — silicon and lasers — have been separate in the past, but IBM can now make them side-by-side, combining optical with electrical circuits on the same chip.

The result of this “breakthrough” could unplug the bottleneck in giant data centers and supercomputers, which have been throttled by data traffic jams and high-cost interconnection systems.

After years of research, IBM says it will be able to manufacture these optical chips in standard silicon semiconductor chip factories. That is important, since the problem slowing down computers today isn’t processing or memory, but the interconnections between them that have become bottlenecks. IBM says it can now move the technology from the lab to “fab,” or commercial silicon chip fabrication plants.

“This technology breakthrough is a result of more than a decade of pioneering research at IBM,” said Dr. John E. Kelly, Senior Vice President and Director of IBM Research. “This allows us to move silicon nanophotonics technology into a real-world manufacturing environment that will have impact across a range of applications.”

Right now, optical components can guide laser light and convert it into electrical signals, which can be processed in silicon chips. But the optical components are bulky and expensive. By embedding the optical components inside the silicon chips, IBM will be able to create data highways that can keep up with the stream of information that has to be processed inside high-speed servers and supercomputers.

“This will help us get past a significant problem in the industry,” said Solomon Assefa, nanophotonics scientist for IBM Research, in an interview with VentureBeat. “When you do a Google search today, it happens inside a big data center, not inside a single chip. The data are stored all over the place, and you have to reach it through fast interconnects.”

The silicon nanophotonics technology can be embedded inside a 90-nanometer chip. Such a chip is easy to make and isn’t as advanced as a state-of-the-art 32-nanometer chip (which has smaller circuitry), but it can be manufactured at low costs and meet performance requirements over the next decade.

Silicon nanophotonics takes advantage of pulses of light for communication, but it is different from other solutions — such as those that Intel has tried to make for years — because it is 100 times smaller.

“We build at the smallest level where these components can direct light,” Assefa said. “the technology is ready for prime time.”

It is just in time for the era of “big data.” Data being created and transmitted over enterprise networks continues to grow due to an explosion of new applications and services. Silicon nanophotonics can enable the industry to keep pace with increasing demands in chip performance and computing power. Today’s optical components have to be chained together in a very expensive solution.

In this era, computers will need to access data that is either a few centimeters or a few kilometers away and move terabytes of data using the light pulses through optical fibers.

IBM began its current approach as a test in 2010. Now it can transfer the technology in a factory, simply by adding a few processing modules into a factory line. The optical components — such as wavelength division multiplexers (WDM), modulators, and detectors — rest side by side with complementary metal oxide semiconductor (CMOS) electrical circuitry (a normal silicon chip).

A single chip could transfer data at 25 gigabits per second, something that requires a chip and a number of expensive optical components today. It can also feed a number of parallel optical data streams into a single optical fiber, or multiplex the streams. That will in turn allow  the communications to scale up to delivering terabytes of data between distant computer systems.

Assefa will deliver a presentation on the subject at the IEEE International Electron Devices Meeting (IEDM) in San Francisco this week.

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

Last week Chinese e-commerce giant Alibaba shocked the web with news that its subsidiaries Taobao (like a Chinese eBay) and T-mall (like Amazon) sold a massive $3.06 billion in product in a single 24-hour period.

Three billion dollars is almost triple the entire 2011 Black Friday sales of e-commerce sites in the United States, and this is a testament to both the growing maturity of the Chinese online market and the central position Alibaba holds in China.

So how do you do $3 billion in sales in a single day? I interviewed an Alibaba spokesperson, who preferred to remain unnamed, to find out.

Almost as many users as Twitter

First of all, it helps to have users — lots of them.

The two sites, Taobao.com and Tmall.com, share a combined user base of half a billion registered users. That’s only about 30 percent of the total Chinese population, which means that Alibaba has a long runway for continued growth as the Chinese middle class continues to grow — and suggesting that a $10 billion day is not out of the realm of possibility in years to come.

Some of the goods for sale on Taobao.

(It also means, by the way, that it might be crazy for Yahoo to sell its stake in Alibaba now. Yahoo still owns 23 percent of the Chinese e-commerce giant after pocketing $7.6 billion U.S from selling almost half its original stake. Down the road, however, it might be like owning a 23 percent of Google, or Facebook … times 10.)

Both Taobao.com and Tmall process payments via Alipay, Alibaba’s payments processor. Alipay has over 700 million user accounts, presumably all with credit card payment information, as of mid-2012.

A cultural sea change

When you’re looking to scale quickly, it helps to be a high-growth company in a high-growth industry in a high-growth country.

But while a huge number of users is necessary, it’s insufficient to generate such massive cash flow. Just as important is a change in Chinese consumer behavior toward e-commerce “increasingly becoming a primary shopping channel,” according to Alibaba. The entire industry has had double-digit growth year-over-year — since 2010 e-commerce in China has enjoyed an average 10 percent quarter to quarter growth rate — and Alibaba is outpacing the industry.

In addition to a general acceptance of e-commerce as perhaps the first option for shopping, Chinese consumers have been increasingly prone to spend big on “double sticks day,” 11/11. Also referred to as “singles’ day,” Nov. 11 has taken on a rough similarity to our Valentine’s Day. Singles try hard to not be single, couples celebrate that they are couples, and probably many more people are simply happy to dogpile any opportunity to shop big and save big, much like Black Friday in the U.S.

Stores galore

Users and demand are great, but you need supply. That is not a problem for Alibaba, as the representative said that Tmall hosts e-commerce operations for 50,000 companies, while Taobao manages online sales and payments and “several million,” most of whom are likely individuals buying and selling much as we see on eBay here.

Of those stores, 10,000 took part in the 11/11 festivities with special sales and promotions. That’s five times the number that participated just last year, which gave Chinese shoppers “access to an even wider range of products.”

Promotion helped too, Alibaba says:

“We also offered new functions, activities and games via the event landing page beginning mid-October and noticed that many online shoppers were visiting the site and browsing through participating stores and available products as well as adding products to their shopping carts.”

Let’s get technical

While I generally like to ask companies for technical details about their server setup, Alibaba was anxious not

to share too much information that might be construed as proprietary, saying only that Alibaba Cloud Computing manages all Taobao and Tmall’s infrastructure from three data centers in Beijing, Hangzhou and California.

It also preferred not to disclose what kind of servers it runs, including operating system and webserver details, but a quick search reveals that Alibaba uses Tengine for webserving, which is a 100 percent compatible fork from the increasingly popular Nginx server, and that Alibaba runs, as expected, on Linux.

Mobile buyers figured prominently in Alibaba’s monster day, with one in four Taobao users accessing the site from a smartphone or a tablet. That compares to one in six last year.

Summing up

A $3 billion day doesn’t happen all the time, even in the world’s most populous country. And it’s a huge proportion of Alibaba’s annual goal for Taobao and Tmall, which was one trillion RMB in 2012. That’s closing in on $15 billion U.S., which is serious money.

Given Alibaba’s trajectory and China’s growth, it would seem that a $4 billion or $5 billion day would not be out of the question in 2013.

photo credit: dcmaster via photopin cc, Autorun！ via photopin cc

Facebook has spilled a few beans just now about its homebrewed data center cooling system, with special tweaks to operating in tax advantaged, weather disadvantaged areas.

Facebook’s Prineville, Ore., data center is its flagship for open-source hardware. As part of its ongoing Open Compute Project, Facebook has been very open on how it keeps its servers cool with a system that only uses outdoor air to ventilate the machines — no air conditioning. The servers there run at a higher-than-normal temperature, but with little or no negative impact on efficiency and hardware lifespans.

However, the same approach wasn’t as feasible for the social network’s data center in Forest City, North Carolina, where each summer the only thing higher than the temperature is the humidity.

“When we started looking at Forest City, the bin weather data suggested that refrigeration might not be required, but ASHRAE 50-year design weather maximums suggested otherwise,” writes Facebook mechanical engineer Dan Lee today on the company blog.

“To try to make the free cooling system work in Forest City, we expanded the server environmental conditions on the high end. Because dry bulb temperatures are warmer in western North Carolina than they are in central Oregon, we set the upper end of the server inlet temperature range at 85°F, instead of at 80°F. And because of the higher humidity in North Carolina, we expanded the relative humidity (RH) maximum from 65 percent RH to 90 percent RH.”

Thanks to climate change and what have you, North Carolina experienced record highs this year, with temperatures peaking at and even exceeding 100°F. But the new system held up just fine, Lee said.

“Despite the record-breaking heat, we didn’t run the DX coils at all this past summer. … When the record hot days occurred, relative humidity was low, allowing the misting system to provide all the needed cooling.”

Amazingly enough, the Forest City data center ended up being slightly more energy-efficient for the summer months of 2012.

For those of you who are fascinated by the subject, the Open Compute Summit is coming up in January and will be held in Santa Clara, Calif.

Top image courtesy of Jolie O’Dell

---

Big Data and Predictive/Real-time Analytics startups: Are you looking to jumpstart development & accelerate market traction? Sign up for the SAP Startup Focus program to receive technology, support, resources and community to help you develop new applications on SAP HANA, a cutting edge database platform. Get started here, and enter promo code “VB2013″ on the form.

---

When Hurricane Sandy took out the data center that hosts Gawker, Buzzfeed, the Huffington Post, and many other sites, those sites went dark. But they weren’t offline for long: Buzzfeed and Gawker both turned to popular blogging service Tumblr (also based in New York) to continue bringing news of the storm and everything else to the public.

While Buzzfeed was able to get back up on its own servers in a few days, Gawker’s network of sites — including Gizmodo, Deadspin, io9, and Jezebel — is still living primarily on Tumblr. What you’ll find among these makeshift sites is the same content you’d normally find there, only with a more liberal approach to posting content.

Some stories don’t have headlines and others are “reblogs” from other Gawker Tumblr sites. The makeshift Tumblrs also feature a cleaner design that brings back memories of what Gawker used to look like before its 2011 redesign.

“All things considered, we’ve done a great job,” Gizmodo writer Leslie Horn told me from a temporary office. “It’s been stressful and disheartening, but I think there’s a silver lining in all of it. We proved to others that even a hurricane can’t stop Gawker.”

The company has done surprisingly well with Tumblr, taking advantage of the flexible platform and letting writers go wild. Some of my favorite posts this week include “Manhattan’s Impending Uptown-Downtown Civil War: Who Will Win?” and the “8 Tips For Tending to a Glamorous Drug Addict After Hurricane Sandy“. Gizmodo even continued to post full product reviews, and Horn set up a Tumblr called Gizmodo: The Movie.

“Tumblr is a largely amateur world,” Gizmodo Editor-in-Chief Joe Brown said. “These guys are professional bloggers. You give them these tools, they are going to blow the average out of the water.”

While the new design carries almost no ads, each site has a prominent State Farm sponsorship message at the top of the homepage, showing that the business side of Gawker is just as resilient as its technical and editorial components.

“Everyone has stepped up really well,” Brown said. “Within five minutes, it was apparent how we were going to handle it.”

Brown said Gawker’s network of sites should be back to normal on Monday, granted everything goes to plan. He noted that Gizmodo has no plans to continue using Tumblr, even for promotional use, as it does with Facebook and Twitter. However, he said it’s always an option.

“We reserve the right to influence and invade every dark corner of the Internet,” Brown said. “We have no plans to use Tumblr going forward, but we always could go that direction.”

While I’ll miss the fun, sometimes crazy content the Gawker team has put together using Tumblr, I can appreciate a company getting back to normal after a damaging and disastrous event.

Dell revealed a big component of its enterprise strategy today, saying it will offer an “active systems infrastructure” that could make enterprise computing more efficient, flexible, and affordable.

The strategy addresses a significant problem of the Internet era: big companies have to deploy new web services and data centers faster than ever. They must build racks and racks of computer server systems, going from zero to 60 miles per hour in a short time. This is a major issue for companies.

Enterprises have to be more agile, rock solid, and streamlined than ever before, said Marius Haas, (pictured), president of Dell Enterprise Systems. And they need a computing infrastructure that enables them to do that.

Dell’s Active Systems Infrastructure is a cookie-cutter platform for enterprises to adopt and get new data center servers up and running in a short time, Haas said, speaking at an event in San Francisco.

In an attempt to make servers more efficient, Dell also revealed a new tiered architecture for storing data in memory in servers. The advance will deliver five times to ten times better memory performance in servers.

Dell has invested in multiple acquisitions and spent billions of dollars to get to this point, said Dario Zamarian, a vice president and general manager at Dell.

A key part of the infrastructure is new server hardware, including the Active System 800, a server rack that Dell says takes 75 percent fewer steps from “power on to production.”

Dell also has a new input-output system aggregator to make sure data feeds into the servers at a higher speed, allowing 2.3 times more computers per rack. The system uses the latest Intel-based hardware that allows for 45 percent better system performance per watt.

The Active Infrastructure family of products and services enables virtual desktop infrastructure, private clouds, and enterprise applications. That means that enterprises will be able to let employees access secure enterprise apps however they want, and the data will be backed up to a private cloud-based data center that protects it from accidental loss.

The challenge is to make a modern data center “scale,” or grow with the ebb and flow of demand from users.

Haas said the active system infrastructure initiative is a critical part of the company’s expansion into the $110 billion market for business computing. In the second quarter, Dell’s server business grew 8 percent, networking grew 39 percent, and storage grew 6 percent. That happened while other big computer companies lost market share. Haas said the cloud business market will be a $110 billion opportunity in 2015.

Dante Orsini, the senior vice president of business development at cloud company iLand, said his company runs seven major data centers and it has had to adapt to the rapid adoption of cloud computing. It had to change the way that it deployed data center infrastructure, mainly by buying ahead of what it needed. It also took a huge amount of talent to deploy each data center.

“There’s a lot of complexity going on,” Orsini said.

With Dell’s new technology, iLand believes it will be able to deploy data centers five times faster and with a lot more self-service than in the past.

“And every time we deploy, it will be done right and precisely,” he said.

That’s a bit of hyperbole, of course, because even though the world’s largest search engine indexes perhaps 50 billion web pages, the Internet itself is more than just Google.

But Google’s millions of servers are certainly “one of the most powerful server networks in the known Universe,” as Google says, and almost certainly the most powerful server network on the planet. And today the company is giving the universe a sneak peek inside with a cool online picture gallery.

That gallery, surprisingly, is short on information, saying little about the actual technical details, such as the probably 2 million servers Google runs, including almost half a million at its Georgia data center. Or how many billions of miles of Ethernet cable the company uses.

But it does have a very cool view of the places your computer has already been via Google. Apparently, while I’m sitting here with my butt in my chair, my MacBook Air may be virtually visiting Hamina, Finland, St. Ghislain, Belgium, and assorted U.S. destinations like Iowa and Oklahoma.

And Google does give us some amazing pictures of the machines that connect our lives.

Check them out:

Image credits: Google

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

The  New York Times has published an in-depth piece today that purports to document how wasteful the Internet industry is, but some critics say the report is misleading.

The industry’s data centers “consume vast amounts of energy in an incongruously wasteful manner,” the Times writes, summarizing the findings of its year-long investigation. “Online companies typically run their facilities at maximum capacity around the clock, whatever the demand. As a result, data centers can waste 90 percent or more of the electricity they pull off the grid,” it said.

The Times said it had commissioned the consulting firm McKinsey & Company to help it analyze energy use by data centers.

No one doubts that today’s data centers hold significant inefficiencies. Companies are in the middle of making large cloud deployments, a transition that’s clearly at only in its beginning. Indeed, we’re debating this question of how green we can make the cloud as part of our CloudBeat 2012 conference on Nov. 28-29.

The issue is a really important one, and in our view, it’s way too much to be undercut by exaggerated reporting. But that’s what critics of the article are saying: That it’s hyperbole. They say the Internet industry is actually becoming much more efficient very quickly and that the article doesn’t address any of its demonstrated strides in this regard.

They say the article is factually misleading in several instances: While the utilization rates of servers in data centers is cited by the NYT as between 7 and 12 percent, the article failed to point out that this statistic is derived from IT data centers, not from state-of-the-art data centers run by the Internet companies such Amazon, Facebook, and Google, who are leading the way in modernization, argues Dan Wood, the editor of CITO Research, a publication written for CIOs and CTOs.

Wood also calls into question whether the low utilization rates are really all that bad. “The fact of the matter is that with very stable workloads it is possible to get high utilization and with variable workloads lower utilization would be expected, so you have room to handle spikes,” he writes at Forbes. “Think of it this way: Roads aren’t 100 percent utilized. The telephone system isn’t 100 percent utilized. They are there when they are needed.”

And while the article calls Internet companies “secretive,” Wood points out that no credit is given to things like Facebook’s Open Compute initiative (opencompute.org), a program in which Facebook shared the design for a green data center so other companies could use it.

Meanwhile, Rich Miller of Data Center Knowledge points out that the Times’ lead anecdote about Facebook’s inefficient servers takes place six years ago. For its first installment of this series, he writes, the Times does a fact-laden job of telling only half the story:

What’s missing is the narrative of how the industry has responded to the challenge of its inefficiency and environmental stewardship. The last five years have seen dramatic changes in the way the largest data centers are designed and operated, as companies like Google, Yahoo, Facebook and Microsoft have vastly improved the energy efficiency of their server farms by overhauling their power distribution systems, using fresh air instead of power-hungry chillers (“free cooling”) to cool their servers, and running their facilities at warmer temperatures. New design schemes for modular data centers have emerged, offering  highly-efficient designs to customers with smaller operations than Google or Facebook.  And we’re even seeing a growing focus on renewable energy, highlighted by Apple’s massive commitment to on-site solar energy and landfill-powered fuel cells.

Criticism has also started on Twitter: “NYT article on data centers is hyperbolic,” tweeted John Raymond Hanger, a former secretary of the Pennsylvania Public Utility Commission. “They use less than 2 percent of power; power modern life & prevent more pollution than they cause.”

Similar criticisms are being left in the comments on the Times‘ story.

We’re looking forward to the rest of the NYT series, where hopefully we’ll get a better idea of where precisely the industry is still falling short, and concrete ways it can improve.

[Image credit: Flickr; IntelFreePress]

SAN FRANCISCO — When Intel warned about weak third-quarter sales last week, it blamed it on slow consumer and enterprise PC sales. But it said that data center computers known as servers were meeting expectations. The reason is that the explosion of tablets, smartphones, and other mobile devices helps drive demand for servers, said Diane Bryant, the head of data center computing at Intel.

Speaking at the Intel Developer Forum in San Francisco today, Bryant (pictured above) said that for every 120 tablets sold, a server has to exist to handle all of the web-based traffic that their users create. She said that for every 20 digital signs in a shopping mall and other areas, like the flatscreens in your local coffee shop, you need one server. And for every 20 surveillance cameras deployed, you need a server to sort through all of the video data.

Servers are pretty universal computers that are used in connected systems, small businesses, communications infrastructure, enterprises, cloud services, technical computing, and mission-critical computing centers like stock markets. As the era of “big data” arrives, companies must deploy big data centers to process it.

Intel is sampling its next-generation E7 and E5 server chips for delivery next year. The low-power server chips are based on Intel’s Ivy Bridge chip designs, which combine graphics and computing on a single chip.

Low-power Xeon E3 series chips arrive in 2013 and are based on the 22-nanometer Haswell core now in development. But a new breed of lower-power Atom-based servers, known as microservers, are also arriving. These microservers have chips with a third of the usual power consumption and they use Intel’s Atom processors. Next year, Intel will launch the 22-nanometer Avoton version of the Atom chip for microservers.

BMW said it will use a new Open Data Center Alliance data center, fueled by geothermal and hydroelectric energy, at a location in Iceland.

On the security front, Bryant said that Intel is embedding Deep Defender security technology into its Xeon server chips. Intel has also launched trusted execution technology for its ecosystem. Intel entered the server chip market in 1997 and has improved performance 10,000 times.

Big data was a $4 billion market last year, Bryant said, and it is growing at 30 percent a year. The data comes from everything, like social networks or radio frequency identification sensors.

Ariel Kelman, head of worldwide marketing at Amazon’s web services division, which hosts web infrastructure for businesses, said that it is working closely with Intel to transform supercomputing from a tool for the elite to something the masses can tap into. Amazon’s Cluster Compute Eight Extra Large Instances are ideal for big data server solutions. They start at $2.40 per hour. You can configure your own top 500 supercomputer with 290 instances at a cost of $73 per hour.

Yelp uses AWS to analyze all of the data from clicks on its online user reviews. NASA uses AWS to calculate telemetry data for the Mars Curiosity rover.

Six Flags uses Intel servers in its amusement parks to monitor crowd traffic via surveillance cameras in its theme parks. It reallocates its staff to the high-traffic areas based on the data from the cameras.

While companies like Amazon and Rackspace mightily tout the power of the public cloud, some large enterprises still aren’t sold. These enterprises like the option of an on-premise private cloud that gives them a bit more control and security.

Palo Alto, Calif.-based Nebula, which has raised $25 million in new funding today, takes serious advantage of these enterprise-level concerns and gives big companies tools to enable less-expensive private clouds. As we wrote about previously, the company takes advantage of Openstack technology and creates hardware appliances that make building and managing low-cost data centers a reality.

Nebula’s team is led by CEO Chris Kemp, who was the former CTO of NASA. Many other folks in the company come from places like NASA, Amazon, Google, and Microsoft. If any team can make it easier for enterprises to improve the data-center building process and make data centers more functional, this might just be the group.

“We’re democratizing what Google and Facebook have done with their data centers,” Kemp told me. “You can use your favorite server vendor and plug those servers into our boxes. You plug in and 15 minutes later, you have Amazon running on premise.”

So what’s the company’s biggest threat? Public cloud and virtualization powerhouses that are already dominating the market. “Amazon and VMware are the two that keep me up at night,” Kemp said.

Nebula is still in beta and has a “couple dozen” customers. Kemp said he hopes to launch with general availability before mid-2013.

The new funding round was led by Comcast Ventures, with participation by Highland Capital Partners, Kleiner Perkins Caufield & Byers, Innovation Endeavors, Andy Bechtolsheim, David Cheriton, Ram Shriram, Harris Barton, William Hearst III, Scott McNealy, and Maynard Webb. Silicon Valley Bank is additionally providing Nebula with debt and credit facilities.

Nebula also previously raised a first round of funding in July 2011, but it has not revealed the specific amount. Kemp told me it was more than $8 million (after I guessed that), but less than this new $25 million round. The company has nearly 60 full-time employees, almost all of them engineers.

Star formation photo via NASA/Flickr

VMware announced on Monday a big, $1.05 billion acquisition of software-defined networking company Nicira.

While VMware is a major player in data center virtualization, its purchase of Nicira is meant to make it a similarly large presence in software-defined networking as well. That’s a fancy way of saying it wants to take what it did to server virtualization and do it to network infrastructure.

Though most of the deal was for cash, VMware also offered $210 million in assumed unvested equity awards, a small price for what is sure to be a significant investment down the line.

Founded in 2007, Nicira was in stealth mode up until this February, but it’s already attracted attention from companies as large and diverse as AT&T, eBay, Rackspace, and DreamHost, which rely on its Network Virtualization Platform. With the technology, companies can scale their network capacity to keep up with demand, saving both effort and money.

Early Nicira proponent and Andreessen Horowitz partner Ben Horowitz applauded VMware’s buy. “VMware today [has created] a clear path to becoming the most important infrastructure company across servers, networks, and storage for the next 10 years,” he wrote in a blog post.

VMware’s move was particularly prescient, Horowitz notes, because it comes ahead of a future where networking and storage infrastructure will be more reliable and far more elastic. “ VMware just took a giant step towards bringing the future forward,” he said.

In an interview with VentureBeat, Horowitz’s partner Marc Andreessen agreed, noting the parallels between Nicira and VMware. “Nicira is to networking as VMware is to servers,” he said. “And a lot of companies are still kicking themselves for not buying VMware.”

But the Nicira buy is a bit different, because, unlike VMware, Nicira is in a league all on its own. “There’s really no other company like Nicira,” Andreessen said.

The move is also a big deal for Nicra co-founder Martin Casado, who has worked on the company’s virtualization technology for over 10 years. “VMware’s purchase is basically a validation of everything that Casado has been working on,” Adreessen said. “Few people believed that you could effectively decouple network software from the server, but that’s exactly what Nicira has done,” he said.

While the boards of directors of both companies have already approved the deal, it’s still subject to regulatory approval. Either way, VMware expects the deal to close before the end of the year.

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

Amazon has released a detailed explanation of its massive EC2 outage this past Friday and Saturday. But even with all the new info, services that went down, including Netflix, Instagram, and Pinterest, have yet to fill in some crucial blanks.

This past weekend, Amazon data centers in Northern Virginia failed after a powerful storm ripped through the area. The company’s AWS service health dashboard slowly updated with details concerning “Power Issues,” but we had very little explanation outside of small updates.

In its new detailed remarks, Amazon claims that it encountered many bugs and small problems that ultimately killed power on its backup generators. The generators ran for a few minutes and then failed, leaving the data centers in the dark. Technicians brought the generators back online in just 10 minutes, but by that point, it would require another three hours to reboot all the servers that had been affected.

Amazon also had a bug that screwed up its Elastic Load Balancers, which help distribute and redirect traffic across different Amazon data centers. Another bug in its Relational Database Service kept a small number of data centers from recovering normally.

We contacted Instagram, Netflix, and Pinterest to get more details that could explain their outages better, but all three services declined to give us access to someone who could talk about their infrastructure. Netflix, for example, should not go down in a single Amazon outage because of redundancies, yet it still failed this past Friday. We’re unsure at this time if Instagram and Pinterest have set up their architectures in a similar manner that should normally avoid outages like these.

Yes, the blame mostly falls on Amazon, but it would be good to have an official explanation from a player like Netflix if Amazon screwed up so badly that it killed whatever spared Netflix from crashing during Amazon’s notable April 2011 outage. A tweet by Netflix cloud architect Adrian Cockcroft points to the Elastic Load Balancers being the real problem. He writes that the company “lost instances in one zone, but lost ELB traffic routing to the zones that were working.”

Instagram also should give an explanation because its outage lasted well into Saturday, whereas Netflix and Pinterest were only down for a matter of hours.

Amazon claims it will repair and retest its data center equipment and software to improve its services. The company’s apology at the bottom of its detailed remarks reads:

We apologize for the inconvenience and trouble this caused for affected customers. We know how critical our services are to our customers’ businesses. If you’ve followed the history of AWS, the customer focus we have, and the pace with which we iterate, we think you know that we will do everything we can to learn from this event and use it to drive improvement across our services. We will spend many hours over the coming days and weeks improving our understanding of the details of the various parts of this event and determining how to make further changes to improve our services and processes.

Let us know in the comments (or send me a note to sean AT venturebeat.com) if you have any further insight into the outage that Amazon didn’t cover in its explanation.

Photo credit:  Dudarev Mikhail/Shutterstock

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

After collecting millions of dollars in tax breaks from having a small office in Nevada, Apple plans to give back to the state with new facilities that could cost up to $1 billion.

Apple’s plans, which were first reported by the Reno Gazette-Journal, include the construction of a data center just east of Reno and a new shipping and receiving office. If everything goes to plan, Apple will get $89 million in tax abatements from the city, county, and state during the next decade, a tax burden reduction of about 79 percent.

The company claims the facilities in Reno will be used to support its iCloud, iTunes, and App Store services. As a user of all of those, I can definitely appreciate more bandwidth going that direction.

“We hope to build Apple’s next data center in Reno to support Apple’s iTunes Store, App Store and incredibly popular iCloud services,” Apple spokeswoman Kristin Huguet said in a circulated statement. “If approved, this project would expand our presence in Nevada and create hundreds of construction jobs over the next year, plus permanent jobs at the data center which will add to our existing total of nearly 400 employees in the state.”

The Nevada data center would be used to support Apple’s data centers in Maiden, N.C. and Newark, Calif., and a planned data center in Prineville, Ore.

Reno photo: jcantroot/Flickr

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

If you’ve ever wanted to see how a single email travels from your account to someone else’s, Google has you covered with its new “Story of Send” site.

While the site appears to mostly be a push by Google to show off its green credentials, it is still cool in that it shows the actual process of something many of us do hundreds of times a week. Email travels from our computer browser over cables to Google’s data centers. There the email is copied, scanned for viruses, and sent back out over cables to the recipient’s email account.

Google Green team member Erin Reilly writes on Google’s blog:

“We’ve included videos and photos throughout the journey so you can explore certain areas more deeply. For example, if you’re curious what data center servers look like, we’ve included some photos. Or you can watch a video to learn about how we purchase clean energy from wind farms near our data centers. And because technology doesn’t always have to be serious, you might find a vampire or two lurking around or uncover other surprises on the journey.”

Check out the “Story of Send” video below:

Looks like some London data-center employees are going to have a server slumber-party during the Olympics this summer. Podtime has been brought on to supply “sleeping tubes” so workers can avoid the Olympics traffic.

Traffic on the road is only one of the issues Interxion’s London data centers will face during the games. Online traffic is sure to swell dramatically as tourists, athletes, vendors and more flock to the U.K. capital. But physical traffic could pose an issue as important technicians try to get from their homes to the servers. So, why not just have them sleep over? That’s where Podtime’s sleeping pods come in.

“[The pods are] not something they’re going to live in for the whole period of the [Olympics],” said founder Jon Gray in an interview with VentureBeat. “They’ll probably just go out for a beer, come back and crash in the pod and the next morning they’d already be there.”

Gray started Podtime just over a year ago, inspired by colleagues in the financial district of London who were tired during the day. He started researching the scientific benefits of power naps, which he believes can be very energizing, if done the right way. So he took a look at a number of different options for sleeping in the workplace. It needed to be portable, small, and roomy all at the same time. Japan had already created whole hotels made out of these pods, which he also researched. Eventually, Gray landed on a cylindrical tube, with square doors that can be stacked.

The tubes can be customized per order to include a Temper-pedic bed, shelves, mirrors, LED lights, power outlets, digital radios, televisions, as well as different door and tube colors. Thus far Gray and his team of five have sold 40 pods, but hope to expand across the U.K. and are even looking for distributors in the U.S.

Check out a gallery of the pods below:

via Data Center Knowledge; photos via Podtime and Interxion

Energy management systems startup Vigilent has raised a $6.7 million round of funding to optimize energy efficiency at data centers using the power of “big data,” the company announced Tuesday morning.

One huge problem companies struggle with when it comes to data center management is how much power those facilities use. Google spotlighted its own energy usage at data centers, claiming its centers are 50 percent more efficient than other companies’. But not everyone has the resources to design and implement innovative cooling technologies.

That’s where Vigilent comes in. The company deploys wireless sensors in data centers and crunches all kinds of numbers to find ways to optimize energy efficiency. Vigilent negotiates a certain temperature with the company running the data center and works to come up with the best possible way to keep the center at that temperature.

Vigilent CEO Mark Housley told me there is an exceptional amount of energy being wasted at these centers. Some of that blame goes to poor design, while some goes to lack of innovation in cooling tech. “The companies that sell air conditioners should be ashamed of what they’re selling,” Housley said.

Housley also explained that Vigilent’s software automates the process of cooling and is smart enough to predict events. For example, during the Super Bowl, it would know to deploy more cooling than usual because of the increased heat generated by busy servers.

Vigilent’s first major round of funding was led by Accel Partners and the firm’s new “Big Data” fund. The company also attracted investments from angels Gaurav Garg and Peter Wagner.

“With proven technology and a significant track record with Fortune 50 companies, we believe Vigilent will have a transformative effect on energy efficiency in critical building infrastructure,” said Rich Wong, Partner at Accel Partners, in a statement.

El Cerrito, Calif.-based Vigilent has 90 deployments throughout the U.S., Canada, and Japan, with customers including Verizon, Akamai, and Informatica. Housley said the company is profitable and will likely double its revenues in 2012. The new funds will be used to move into new regions and expand its workforce.

Chilly data center: Supri Suharjoto/Shutterstock

Pop star Will.i.am endorsing Megaupload

In putting the kibosh on MegaUpload, federal prosecutors arrested seven men and froze millions in assets. With their funds in limbo, MegaUpload can’t pay the costs to the companies that store its data, and a letter from the federales says that these companies, Carpathia Hosting Inc. and Cogent Communications Group Inc., can begin deleting the files stored with them as early as this Thursday.

This can’t be welcome news to the people who claim to have uploaded important and perfectly legal files to MegaUpload, and are currently in the process of trying to sue the FBI for seizing their data. To be fair, that group of affected users is spearheaded by a pirate site, which doesn’t exactly lend credence to the idea that there was a large group of users relying on MegaUpload for legit purposes.

Beyond the collateral damage to MegaUpload users, the loss of data might have a more important impact, the destruction of evidence. MegaUpload’s attorney, Ira Rothken, argues that the files stored with these third-party companies may prove valuable in his defense and that it’s in the best interest of all parties to ensure they don’t disappear. “We’re cautiously optimistic at this point that because the United States, as well as Megaupload, should have a common desire to protect consumers, that this type of agreement will get done,” he told the AP.

The FBI copied some data during its seizure, but left the bulk untouched. The government, after seizing MegaUpload’s assets, hasn’t stepped forward yet to assume the costs of keeping this data while the trail moves forward. It’s a reminder that criminal investigations haven’t caught up with the reality of today’s data centers.

Remember that in June of last year, the FBI knocked dozens of innocent sites offline when it seized servers at a hosting facility in Reston, VA. Perhaps the big takeaway is that when it comes to policing the criminal use of data, there is a good chance of collateral damage.

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

How do you sex up an input-output system that will blow away data center bottle necks? Have Verne Troyer announce it at a groovy DEMO party with drink serving go-go girls!

Fusion-io, announced its new auto commit memory system, which allows a company to execute one billion input-output commands per second (called IOPS by techies) in a rack of eight servers. Previously, an I/O rate that high would require a roomful of servers, the company said. This plays into Fusion-io’s goal of shrinking the data system and making it as efficient as possible. Fusion-io chief scientist and Apple cofounder Steve Wozniak described it as, “least costly, most efficient, fastest operating.”

The company, which went public in June 2011, has grown based on its ability to deliver newer, faster ways of moving data from storage to server CPUs, as people learn more about the information at their finger tips. Fusion-io’s flash chips are more reliable than traditional solid-state disks (SSDs) and are much faster. The company released its first chip, at the DEMO conference in 2008, and released its newest iteration, the ioDrive 2 in October 2011. The ioDrive 2′s purpose is to speed up web page load time, while not overheating your servers.

For David Flynn, chief executive of Fusion-io, the company’s main purpose is to deliver useful products to the cloud computing generation, and move away from the outdated hard drive storage still used in many data centers.

The New York Stock Exchange uses Fusion-io’s technology. Not coincidentally, it is also the exchange Fusion-io stock trades on.

This new technology is called auto commit memory, which, according to Fusion-io’s blog, “is a new memory type that uses the underlying flash to present a persistent memory directly to applications.” It’s an extension to Fusion-io’s ioMemory architecture, and promises to significantly reduce latency and overhead in servers when transferring data.

For the press conference, the company set up a rack of eight HP ProLiant DL370 servers, each equipped with eight ioDrive2 Duos, and displayed a meter on-screen that they said showed the rack was executing one billion I/Ops. Fusion-io was able to execute a million in 2009. The next year it did the same with HP’s servers. The auto commit memory extension will be available in April of this year.

“The company culture starts from the beginning,” said Wozniak at the event. “One of my big jobs was recognizing what was beautiful about the various products. What’s in that rack there, last time we were here, would have filled this room.”

The company is able to achieve this by making chips that speed the process of moving data from short term to long term memory by putting data closer to the processors which will call for it first. The average processor utilization today is less than 20 percent, says Flynn. So by making memory faster, he says, Fusion-io’s technology can potentially improve server performance fivefold, by getting CPU utilization closer to 100 percent.

Wozniak, Flynn and Fusion-io chief marketing officer Rick White announced the product at the DEMO event, Enterprise Disruption: An Evening of Change and Innovation in San Francisco.

Check out the announcement video with Verne Troyer above. And don’t miss our gallery of photos of Woz and special guest Leonard Nimoy.

Sept. 9 - 10, 2013
San Francisco, CA

Early Bird Tickets on Sale

Better grab an umbrella, because it’s about to get real cloudy. A new study by Cisco estimates that global traffic generated by cloud computing services will increase a staggering 12 times by 2015 compared to cloud traffic in 2010, while data center traffic will increase at a less-showy-but-still-impressive four times by 2015.

As one of the biggest cloud computing providers in the world, Cisco has an interest in tracking and estimating how big its services and clientele could be in the next few years. The company’s Global Cloud Index, from which the eye-popping traffic numbers originate, is based on measured data center traffic, analyst projections, and academic studies.

Cisco has tracked and estimated future data center traffic for several years, but this is the first time it has looked at purely cloud-based traffic. The study confirms what many experts expect: that cloud service adoption will greatly accelerate in the next several years as more companies and consumers realize the potential of cloud solutions and storage. Global cloud traffic will increase from 130 exabytes (EBs) in 2010 to a projected 1.6 zettabytes (ZBs) in 2015, a 12-fold increase that will take the cloud to the “zettabyte era,” according to the study.

“The study was key learning for us,” Thomas Barnett, Cisco senior manger for strategic communications, told VentureBeat. “Many users don’t realize that sending a file doesn’t generate an equal amount of traffic because the file has to go through servers and such. Sending a 5-megabyte file, for example, actually generates around 50 megabytes of traffic.”

In terms of pure data center traffic around the world, traffic is projected to go from 1.1 ZBs in 2010 to an estimated 4.8 ZBs in 2015, four times the amount. 4.8 ZBs is a hard-to-imagine number, so Cisco has quantified it to equal 66.7 trillion hours of streaming music at 160 kbps, 15.5 trillion hours of standard-def web conferencing or 4.8 trillion hours of online streaming 720p HD video.

Barnett also said the study indicated the average server will go from supporting 1.4 workloads in 2010 to 2 workloads in 2014. By that year, Cisco expects more than 50 percent of all workloads will be processed in the cloud instead of in local data centers, something that would indicate an enormous shift of resources and the way IT is managed for the enterprise.

The most notable reason for shifting from local data centers to cloud data centers? Cold hard cash. As an enterprise uses more servers operated by cloud companies, the cost per server greatly decreases. When a company reaches the 1,000-server mark, the cost differential between using local data centers and cloud data centers is nearly 50 percent.

Check out the accompanying infographic for more details from the Global Cloud Index study:

CloudBeat 2011 takes place Nov 30 – Dec 1 at the Hotel Sofitel in Redwood City, CA. Unlike any other cloud events, we’ll be focusing on 12 case studies where we’ll dissect the most disruptive instances of enterprise adoption of the cloud. Speakers include: Aaron Levie, Co-Founder & CEO of Box.net; Amit Singh VP of Enterprise at Google; Adrian Cockcroft, Director of Cloud Architecture at Netflix; Byron Sebastian, Senior VP of Platforms at Salesforce; Lew Tucker, VP & CTO of Cloud Computing at Cisco, and many more. Join 500 executives for two days packed with actionable lessons and networking opportunities as we define the key processes and architectures that companies must put in place in order to survive and prosper. Register here. Spaces are very limited!

The Open Compute Foundation, announced today, will allow anyone to access the designs and specifications for Facebook’s homebrewed and highly efficient data center hardware and will provide structure for the project.

To join, hardware designers and hackers need to sign an agreement on the Foundation’s site, find an aspect to work on and sign an open licensing agreement to do that work. Designs will be voted into the official project based on merit.

The goals are to make data center energy usage more efficient, to make server repairs faster and fewer and to figure out better ways to serve data at massive scale, e.g., billions of users, what is known as “human scale.”

“This is not going to be a community that’s going to write wish-lists,” said Frank Frankovsky, one of the founders of the Open Compute project at a summit in New York City this morning. “We’re publishing not only specifications but also source files.”

“When we first launched this project back in April, people thought we were crazy,” said Frankovsky.

Now, however, Facebook’s groundbreaking work in data center efficiency and server redesign will be open for all to hack on and improve. As an example, Frankovsky mentioned Facebook’s newest data center in Sweden, a hydropowered facility that he said is “our greenest yet.” That design will be open sourced, its specs published and available for anyone to manipulate and possibly enhance.

Right now, you can go to the Foundation website, login with your Facebook and Github accounts, and start checking out the components that make up Facebook’s data centers, from the motherboards to the chassis — even Facebook’s completely redesigned power supplies.

“Let’s build this together,” said Frankovsky. “If we start sharing our ideas, the pace of innovation is going to increase rapidly. … Let’s start focusing on the environment and the efficiency of this.”

Asus, Intel and AMD are involved as partners and will be publishing specs and source files, which hackers will then be allowed to modify and submit back to the project. Dell is also a major partner for the project.

Facebook’s Open Compute project has inspired the interest of the hardware hacker and open-source communities, individuals and groups that Facebook said “are passionate about making strong technical contributions to defining and delivering the most efficient server, storage and data center designs.”

Open source is a huge part of Facebook’s culture. Pretty much since its inception, the company has both used and created or contributed to open-source software projects. In a recent conversation with Facebook open-source software lead David Recordon, we talked about how open-source is simply part of Facebook’s DNA in a way that’s rare for a Silicon Valley startup.

“I think it’s pretty clear there’s no question about whether companies should be using open-source software or not,” said Recordon to VentureBeat back in August. “That was answered over the past decade. The question now is about open hardware. Many of the things that we have today for open-source software we don’t have for hardware and standards.”

Amir Michel, who heads up Facebook’s open-source hardware efforts, also told us in an interview on Facebook’s data center redesign, “A lot of the tools aren’t there yet,” said Michael. “If someone wants to make a change to one of our circuit boards, it takes hundreds of thousands of dollars to get that package. The average hacker doesn’t have that. Most of the contributions so far come from other large companies. We’re hoping to change that in the future so a guy in his garage can design a motherboard.”

Stay tuned for more news on open-source hardware as the Open Compute Foundation begins accepting projects from new contributors outside Facebook.

Check out DevBeat, VentureBeat’s brand new channel specifically for developers. The channel will break relevant news and provide insightful commentary aimed to assist developers. DevBeat is sponsored by the Intel AppUp developer program.

---

Big Data and Predictive/Real-time Analytics startups: Are you looking to jumpstart development & accelerate market traction? Sign up for the SAP Startup Focus program to receive technology, support, resources and community to help you develop new applications on SAP HANA, a cutting edge database platform. Get started here, and enter promo code “VB2013″ on the form.

---

ioCache integrates flash memory and caching, or saving data in the cache for later use, to free up otherwise redundant actions in servers. This allows for faster data communication and more room for virtual computers on one server.

The product was born out of Fusion-io’s purchase of IO Turbine, a company specializing in caching for “virtual environments,” or multiple computers living on one physical device. The company purchased IO Turbine for $95 million in early August. The purchase was intended to further Fusion-io’s data center optimization goals.

Now, ioCache is allowing enterprises to do just that: squeeze as much virtualization juice out of their servers as possible.

When more virtual computers live one one server, costs go down even in the physical realm. When we think of virtualization, most people don’t consider the tangible toll of server upkeep. Cooling, powering and managing servers requires man hours, which can be expensive, not to mention how expensive servers themselves are. These costs are reduced when companies are able to downsize server purchases simply by adding more virtual computers to one closet.

According to Fusion-io chief executive David Flynn, the company was created for the era of cloud computing, replacing outdated computer storage technology. With “cloud” on the rise, the virtualization space getting attention with companies like Hotlink receiving funding for virtualization managment. Competitors  associated with Fusion-io’s enterprise storage business include EMC, Hitachi Data Systems, and NetApp.

Jay Phillips, Fusion-io’s vice president of virtualization solutions, will present ioCache at the VMWorld, a virtualization conference, this Thursday.

Fusion-io was founded in 2006 and raised $115.5 million before its initial public offering of $233 million in June. The company’s founders David Flynn and Rick White are accompanied by Fusion-io chief scientist and Apple co-founder Steve Wozniak.  Fusion-io launched its first solid state memory disk at the DEMO conference in Fall 2008.

Come to DEMO in Silicon Valley Sept. 12-14 to see new product launches from a worldwide selection of disruptive companies. Register Today and take advantage of our special VentureBeat Partner rate of $995.00.

It’s one thing to open-source the code for your app — that’s a simple matter of mashing a button on Github. But how do you really open-source hardware?

Think about that: Facebook committed to open-sourcing the infrastructure of its data centers through the Open Compute Project, which launched back in April. But there’s more to maintaining an open-source project than just releasing data into the wild. You also have to accept contributions from other members of the community.

So how do you accept a patch for a motherboard? Or an improvement to a power supply?

This was just one of many challenges facing Amir Michael and the rest of Facebook’s open-source hardware team as they began redesigning the company’s servers and data centers. And to be frank, it wasn’t even the most challenging problem they’ve faced so far.

How Open Compute began

Michael, a former Googler, told VentureBeat that when he first came to Facebook, “I knew a lot about servers and data centers.” Not only did he understand the architecture of a network of servers; he even had hands-on experience as a data center tech, where he often worked until his hands were raw from repairing downed machines and replacing faulty components.

At the time Michael first came on board at Facebook, he said, “The way Facebook was scaling was tremendous. We were buying servers from HP and Dell, leasing server space from Data Realty Trust.”

But Michael had an inkling that one of the most significant companies of the decade might not actually have been handling its data in the smartest, most efficient way. “I did a little analysis,” he said. “I went to NewEgg.com and put together an equivalent server, and it was about the same price, even though we were buying in these huge volumes.

“The business model didn’t make sense.”

After realizing that Facebook wasn’t doing itself any favors buying stock servers in huge quantities, Michael started investigating how the servers were cooled and powered. “I realized there was a lot of inefficiency there, too,” he said. “We looked at how to improve it. With optimizing the data center and ignoring the servers, you get some efficiencies, and you get some efficiencies by optimizing the servers and ignoring the data centers. But you get the biggest benefits if you optimize both.”

In a nutshell, that’s how the Open Compute Project was born.

At the outset, Michael and the Facebook team tried to work with their existing hardware providers. “The vendors’ responses to the changes we wanted to make were lukewarm,” Michael said. “They offered to do a bunch of other things that weren’t too useful for us. They wanted us to buy what their other customers were using, but those modified machines weren’t as extreme as the customizations we were considering.”

Redesigning the server

From that point forward, Michael, Facebook’s manager for hardware design, started tearing apart every assumption about how servers were supposed to be built.

“We looked at why things were done the way they were, and it always came down to legacy. Challenging legacies and starting from scratch was the most innovative thing we did in the project,” he said.

For example, in Facebook’s new server design, the way power is delivered to the microprocessor is entirely different. The team took out transformations and distribution mechanisms and changed the power supply itself. Even the power cords and power strips have been entirely re-engineered, and the servers themselves were designed to be built and maintained without any tools.

In fact, Michael said the serviceability of the server was one of the team’s most important innovations. “When you have tens of thousands of servers, they break on an hourly basis. The hard drives fail, the memory fails. Our data center technicians are responsible for maintaining the servers. They spend their whole day installing new cabinets, new hard drives, etc. We wanted to make their jobs as easy as possible and a lot more efficient. We didn’t require any tools to assemble the servers, and most components are two to 10 times faster for basic service functions than on an average server.”

To test out this aspect of efficiency, Facebook had a prototype build party, which goes down in our book as one of the nerdiest ways to have fun on a Saturday night. “We let a bunch of engineers build the servers, we had pizza and beer, and we had a competition to see who could build a server the fastest,” said Michael. “A data center tech got it built in eight minutes.”

Facing resistance

When Michael was done redesigning the most fundamental aspects of the server, however, he didn’t get an initial enthusiastic response from a few key audiences. Engineers at Facebook who had to do work on the servers were “skeptical,” he said, and even the new vendors were “hesitant.” Facebook’s management took some convincing, as well.

“In general, the resistance to change, getting people to accept a new architecture, was our biggest challenge,” he said. “Getting people to be open to trying something new was hard.”

That’s especially true for big hardware changes. Making radical software changes is, by contrast, cheap and easy. “With hardware,” said Michael, “you need a lab, new hires, prototypes. It requires several million dollars worth of investment. To their credit, Facebook management’s willingness to invest in this fringe project speaks to their ability to take big risks and allow for innovation to occur.”

Those big risks involved trip after trip to Taiwan to work with new manufacturers, bringing a mechanical engineer in-house, and drafting between 50 and 60 pages of specs for the new servers. “Doing design on a white board is one thing, but figuring out the details is where you can stumble, said Michael, “especially when everything you’re doing is customized and entirely new from the ground up.”

Open-sourcing hardware

Finally, the Facebook team is still trying to figure out how to make the Open Compute Project truly open source by accepting contributions from the hardware hacker community.

“A lot of the tools aren’t there yet,” said Michael. “If someone wants to make a change to one of our circuit boards, it takes hundreds of thousands of dollars to get that package. The average hacker doesn’t have that. Most of the contributions so far come from other large companies. We’re hoping to change that in the future so a guy in his garage can design a motherboard.”

Michael continued to say that with the right software, that garage hacker could be making contributions as innovative as anything coming from a lab at HP or Dell. Currently, even the software used to design hardware is prohibitively expensive. But this is code — invisible, intangible ones and zeroes — and there’s no reason it shouldn’t be free.

Facebook wants to work with software vendors on free licenses for Open Compute Project contributors. The company is also considering working with other corporations and organizations (such as governments and large universities, which have similar computing needs) to create new, open-source software programs for hardware design.

Another prohibitive aspect is prototype creation. A typical prototype server might cost between five and 10 times more to build than a production server, so even garage hackers might need to get some kind of financial backing for those projects.

The philosophy of open-source at Facebook

We asked Michael if he had any ideological qualms about being an open-source guy at a proprietary software company. “As the guy who builds the infrastructure, I’m disconnected from the software that runs the site. It’s not a dilemma I experience on a daily basis,” he said.

But he continued, “It’s natural in an environment where companies are trying to remain profitable to keep some pieces of innovation to themselves. But they also need to be able to share and engage with the community. If you think about our business model, it’s about providing a valuable service to our users. The infrastructure we use to do that wasn’t a key piece of the business model. Our advantage is the product, not the servers. It’s not a core piece of IP.”

Also, Michael said, “Engineers are social beings, too, and they like being able to talk about the things they’re passionate about. And when you share information, you get benefits. You get feedback from other people about better, cheaper ways to do things.

“If you look at how Facebook was built, it uses a lot of open-source software. We’ve contributed back a lot in the software world, but we haven’t contribued back to the hardware world yet. No one has. But if we do that, maybe other companies can use the same kind of infrastructure. They don’t have to waste energy, and they don’t have to go through the same development process we did.”

Sharing information with universities has been particularly fruitful, Michael told us. “They have interesting solutions, but they don’t have enough data about real-world problems. They don’t know how industries operate. So by sharing information about our workloads and configurations, we get a lot of interest from universities.”

“Then there’s the environmental impact,” he said. “If we share these best practices, we’re hoping that other people can adopt it and have an impact on the environment as well.”

Check out DevBeat, VentureBeat’s brand new channel specifically for developers. The channel will break relevant news and provide insightful commentary aimed to assist developers. DevBeat is sponsored by the Intel AppUp developer program.

---

Big Data and Predictive/Real-time Analytics startups: Are you looking to jumpstart development & accelerate market traction? Sign up for the SAP Startup Focus program to receive technology, support, resources and community to help you develop new applications on SAP HANA, a cutting edge database platform. Get started here, and enter promo code “VB2013″ on the form.

---

Sentilla’s software uses physical equipment and virtual cloud IT infrastructure to improve data center efficiency by discovering, monitoring, analyzing, planing and automatting data center workloads. Sentilla’s energy management platform is currently in its sixth generation. Sentilla’s service is subscription-based and the company says it currently helps customers reclaim around 18 to 20 percent of their power capacity.

If the terms “IT infrastructure” and “IT department” have the tendency to make your eyes glaze-over, this might make you sit up: Sentilla has patent-pending Virtual Meters that compute the power consumption of unmetered assets. These meters perform “what-if” analysis to predict future performance and then track application performance in both virtual and dedicated environments. This information has been hard to track historically, making these meters very valuable.

“We’re leading the charge to turn IT upside down and deliver more services using less resources,” said Mike Kaul, CEO of Sentilla, in today’s press release. “Sentilla is the only company that can provide the diagnostic insight to drastically reduce data center costs AND increase company revenue as well as profitability. This funding is a validation of what we’ve been seeing in the market and hearing from our valued customers, and provides us with the resources to accommodate growing demand.”

Today’s financing round is lead by SingTel Innov8 Ventures with participation from Sentilla’s existing backers ONSET Ventures and Claremont Creek Ventures. Founded in 2003, this is the company’s third round of institutional funding. They have raised a total of $26.3 million. Sentilla will use today’s funding for “sales and marketing expansion.” The company is headquartered in Redwood City, CA with a branch in London.

Sentilla competes with Power Assure, Modius, VMWare and SynapSense (used by Facebook). HP recently jumped into the game with a “Discovery and Dependency Mapping Service for Data Center Transformation.” Besides the Virtual Meters, what sets Sentilla apart is the company’s experienced team. CEO Kaul has an impressive resume. A former United States Air Force pilot, he has more than 20 years of experience in software. Before Sentilla, Kaul was president and CEO of diCarta. Prior to that he was Charman of BuildPoint, after being CEO of HotDispatch, and the former VP and GM of Oracle’s New Technologies devision.

Sentilla has won the Cool Vendor award from Gartner in data center infrastructure management. The company was also included in GigaOM’s Structure 50 list of the top cloud computing and infrastructure companies. The company was co-founded by CTO Joe Polastre who was featured in the 2009 Silicon Valley/San Jose Business Journal 40 Under 40 award and named one of BusinessWeek’s Best Young Tech Entrepreneurs. Polastre has held software development and product manager positions with IBM, Microsoft and Intel.

Sentilla will be showcasing its cloud analytics tool at VMworld in Las Vegas, starting August 29. You can read more about the tool on the company’s blog.

(image courtesy wiki.engr.illinois.edu)

The company provides a single sign-on for access to servers in addition to management solutions. These solutions include the ability to turn off potentially compromised machines, restrict user privileges, and monitor all activity on the servers.

“A lot of breaches come from insiders who have the keys to the kingdom. We eliminate the keys to the kingdom,” chief executive Tom Kemp told VentureBeat in an interview.

Kemp explained that having multiple sign-ons for various applications on the server is a vulnerability. For instance, firing an employee requires the disabling of his user names and passwords. But this leaves room for access points to fall through the cracks. You may have removed their Google Apps access, but what about the Salesforce account you forgot about?

If there is only one sign-on, however, you can easily retract server access from Centrify’s hub. Okta, which also recently announced a round of funding, is a direct competitor to Centrify’s cloud security offerings.

But what Centrify is most concerned about is how many different devices can now access server information.

“The data center of the future is very heterogeneous and very hybrid,” said Kemp.

He means smartphones and tablets are quickly entering the enterprise as viable ways to distribute data. But these devices are easily compromised. Not everyone password-protects their mobile devices, and they are often forgotten on a train seat or in a restaurant. This pokes an immediate hole in the security of that company’s server infrastructure — cloud or on-premise.

The funding was led by Index Ventures.

Centrify is using some of the funds to build a product combating this vulnerability. The product will allow server managers to deactivate any mobile device associated with a company, in the same way that they can deactivate computers and servers.

While simple, this could bring some peace to worried managers as more and more devices store proprietary information in the cloud.

But while a single sign-on is a good way to manage security, a watchful eye still needs to be placed on existing user activity.

“We do have an auditing solution that provides, in effect, a security camera on the servers and logs all activity done on that server,” said Kemp.

That security camera can be set up to send alerts to a manger and can even be programmed to shut down all systems when a potential security threat is identified. This latter solution could be more effort than it’s worth given the possibility of false alarms.

Another flaw is the inability to detect security threats during work hours. Some threats may come at night and are easily detected. But what about the ones that occur during the normal pace of work? Many servers are exporting and importing data regularly. And if what Kemp says is true, and many breaches come from the inside, it is even harder to detect usual operations versus threats.

According to Kemp, Centrify cannot identify those breaches at this time.

Centrify will also use the funding to expand internationally. Kemp is particularly excited about leveraging Index Ventures’ international presence and believes the firm will play a role in Centrify’s expansion.

Currently, the company has over 3,500 customers including Research in Motion, Pfizer, and the U.S. Army. It has 150 employees, headquartered in Sunnyvale, Calif. Thus far, Centrify has accumulated $52 million in funding from Index Ventures, Mayfield Fund, Accel Partners, INVESCO Private Capital and Sigma Partners.