Facebook announced that they are building their own datacenter and hardware products from the ground up. This includes building their own servers, storage, and infrastructure that complements their “home grown” approach to their software. Facebook is now scaling around 600 million users which includes over 100 million mobile users.
This move is a notice to all the “old school” hardware vendors. They are taking the Google playbook of designing their own systems and hardware and opening it up. Very innovative and compelling for developers and hardware hackers.
Facebook’s announcement is playing the open source card but lets remember there are two kinds of open source: 1) Linux and Hadoop – a community of folks get together and collectively build a product verses 2) Sun, Java, IBM, and now Facebook – we built a product from scratch and now “you” can use it. In scenario 2 there is no input into the product by the community.
Releasing Open Hardware – Open Compute Project
Facebook is announcing the Open Compute Project in conjunction with HP to get a “developer” environment out there. Its goal is to share their efforts that can help other companies looking for the kind of efficiency that Facebook is getting.
This announcement highlights my blog post that I wrote highlighting the Fusion-io IPO analysis.
Inspired by the model of open source software, we want to share the innovations in our data center for the entire industry to use and improve upon. Today we’re also announcing the formation of the Open Compute Project, an industry-wide initiative to share specifications and best practices for creating the most energy efficient and economical data centers.
As a first step, we are publishing specifications and mechanical designs for the hardware used in our data center, including motherboards, power supply, server chassis, server rack, and battery cabinets. In addition, we’re sharing our data center electrical and mechanical construction specifications. This technology enabled the Prineville data center to achieve an initial power usage effectiveness (PUE) ratio of 1.07, compared with an average of 1.5 for our existing facilities.
Everyone has full access to these specifications, which are available at http://opencompute.org/. We want you to tell us where we didn’t get it right and suggest how we could improve. And opening the technology means the community will make advances that we wouldn’t have discovered if we had kept it secret.
Facebook Becoming A Tech Bellwether, and Its Infrastructure is a Megatrend.
All the top industry insiders are watching two major bellwether indicators that map the future growth of IT infrastructure industry – Cloud Scale (low latency storage/compute/networks) and Big Data (Hadoop). SiliconANGLE follows these two areas in depth. We follow Facebook extensively both from a technical angle and business angle. Facebook in fact is an IT indicator for future growth in the IT sector. For the folks not following this closely let me explain: Facebook runs one of the most comprehensive and innovative clouds in the world all on their own “open source and home grown” software.
Facebook is building the model for what most if not all Private and Hybrid clouds will look like. Companies like HP, Dell, and Fusion-io supply key hardware to Facebook. Said another way, the Facebook opportunity (hence impact to vendor sales) is not about how much tech vendors sell to Facebook, but instead how much tech they sell to infrastructures that will look like Facebook – hello Open Compute Project and hybrid clouds now. Facebook is building a future platform while operating at scale.
This is the objective of *all* top infrastructures that will migrate to hybrid and private clouds over the next 5 years. This megatrend is one of the fastest growing markets in the IT enterprise space.