Facebook’s IPO Filing Reveals Technology Details

One of the most anticipated tech IPO filings ever revealed what everyone already knew: Facebook is stinking rich.

The filing disclosed a number of statistics including its 2011 revenue of $3.7 billion, CEO Mark Zuckerberg’s $1.5 million compensation and just how active its users really are: More than half of its 845 million monthly active users visit the site daily.

But aside from some of its jaw-dropping financials and usage stats, the IPO filing provides a peek into Facebook’s technology ethos and infrastructure.

Its infrastructure, for one, is exceptionally fast, processing thousands of pieces of data in less than a second. Fast is key for Zuckerberg.

As Zuckerberg described in an open letter within the filing, at Facebook’s core is a strong dedication to technology innovation and moving quickly, a sentiment echoed by Facebook CIO Tim Campos in an interview last June. In his letter, Zuckerberg describes Facebook’s culture and management approach as “the Hacker Way”:

“The word ‘hacker’ has an unfairly negative connotation from being portrayed in the media as people who break into computers. In reality, hacking just means building something quickly or testing the boundaries of what can be done.

“Hacking is also an inherently hands-on and active discipline. Instead of debating for days whether a new idea is possible or what the best way to build something is, hackers would rather just prototype something and see what works. There’s a hacker mantra that you’ll hear a lot around Facebook offices: ‘Code wins arguments.'”

In 2011, Facebook spent $388 million on research and development—11 percent of its budget. That’s up from $144 million in 2010 and $87 million in 2009.

Facebook’s Fast Infrastructure

According to the filing, loading a user’s home page typically requires accessing hundreds of servers, processing tens of thousands of individual pieces of data and delivering the information selected—which happens in less than one second.

Another example of speed: Facebook says it uses a proprietary distributed system “that is able to query thousands of pieces of content that may be of interest to an individual user to determine the most relevant and timely stories.” Facebook says this is delivered to the user in milliseconds.

Facebook Patents and Intellectual Property

In the filing, Facebook revealed that it has 56 issued patents and 503 filed patent applications in the U.S. These patents range from technologies that connect the relationships between data elements, including images, GPS information, search results, and specific and aggregated data.

Zuckerberg is listed as the inventor or co-inventor on at least seven of the issued patents. But the company doesn’t insist that it invent everything: Last year, Facebook spent $51 million acquiring patents, which is up from $33 million in 2010.

A few more Facebook creations: intellectual property that includes a set of search indices, query processors and caching systems to process queries about the Social Graph; a data warehouse infrastructure called Apache Hive that “provides tools to enable easy data summarization, ad hoc querying and analysis of large datasets,” according to the prospectus; and “HipHop,” which transforms PHP source code into optimized C++ code, it says.

Storing All That Data

Facebook collects a lot of information from its 845 million users. But to get an idea of just how much information it stores, Facebook disclosed that in photos and videos alone, the number is upwards of 100 petabytes—that’s double all written works of humankind in all languages.

To store all this information, Facebook built a number of storage and serving technologies, including one called Haystack. Haystack, which Facebook developed in-house in 2009, is a custom-built file system solution that allows it to reduce its dependence on hardware.

Last year, Facebook also constructed in Prineville, Ore. its first company-owned data center, but it doesn’t underestimate the power of the cloud. Campos addressed Facebook’s stance on cloud last year, saying, “Facebook is really a different planet. From an IT perspective, Facebook grew up with a different stack of IT available to it. We’re a very next-gen tech company from the way we scale our site to everything else.”

Campos also noted that the majority of its enterprise applications are SaaS-based. As a result, Facebook’s operations team is relatively small.

Kristin Burnham covers consumer technology, social networking and enterprise collaboration for CIO.com. Follow Kristin on Twitter @kmburnham. Follow everything from CIO.com on Twitter @CIOonline and on Facebook. Email Kristin at kburnham@cio.com