As data volumes have soared along with its growth, Facebook has built a dedicated network to manage the huge flows of machine-to-machine M2M traffic between its data centers.
The rise of artificial intelligence, and the GPU computing hardware that often supports it, is reshaping the data center industry’s relationship with power density.
Facebook is beefing up its high performance computing horsepower with Big Basin, an AI server powered by eight NVIDIA GPU accelerators. Big Basin was introduced at today’s Open Compute Summit.
Facebook today introduced Bryce Canyon, a new storage unit that adds density by inserting 72 disks vertically into a 4U enclosure, boosting Facebook’s storage capacity. (Photo: Facebook)
As it approaches 2 billion users, Facebook is hitting a new phase of growth, which includes building even larger data centers, and more data centers on each cloud campus.
Microsoft, Google and Facebook are accelerating their data center expansion in Iowa. The three companies are investing nearly $7 billion in their cloud campuses in the state.
To scale up its Mobile Device Lab, Facebook built custom racks to house 2,000 smartphones for application testing in its Prineville data center. Here’s a look at the design process for this offbeat challenge.
The Big Sur GPU-powered server is the key to Facebook’s bid to create a smarter newsfeed for its 1.6 billion users around the globe. We see it in action at Facebook’s Oregon data center.
Machine learning and artificial intelligence have arrived in the data center. This trend is changing the face of the hyperscale server farm, as racks begin to fill with ASICs, GPUs, FPGAs and supercomputers.
If virtual reality becomes widely adopted, it will require a major expansion of the world’s networks and data centers. We look at the infrastructure implications, and how Facebook and the telecom sector are preparing for a VR future.