Monday, November 17, 2014
Facebook's newest data center may impact hardware organization
Facebook's latest data facility, opened recently in Altoona, Iowa, is notable for being the social media giant's first center powered by 100 percent renewable energy. The new data center is also notable for serving as the test run for Facebook's latest and greatest scalable network architecture. Utilizing a data center fabric approach that favors pods over cabinets, this innovative design strategy has the potential to lead the direction of data center design in the future. The construction of Facebook's latest data center also reflects how facilities should remain forward-facing when designing their own network architecture.
Facebook deploys fabric
Seeking to overcome the limitations of organizing the network into clusters, Facebook's latest network will instead connect hardware throughout the building to one, high-performance network, according to Data Center Dynamics. This network is organized into uniform pods, with each pod containing 48 server racks. Traffic for the entire building is routed through four "fabric switches," an architecture that encourages uniform connectivity throughout the building. Brice Towns, the new data center's site manager, remarked that the data center set a record for building construction at a new site. This rapid construction was helped, in part, by the architectural flexibility allowed by Facebook's fabric strategy.
Disruption is always around the corner
Fabric architecture for data centers is liable to gain momentum if Facebook's newest facility is a success. This trend would pose problems for manufacturers, like Cisco and Juniper, whose revenue streams depend on the core-focused infrastructure that still acts as the data center industry's status quo, said Wired. The scalability of the core/tier structure has been deemed insufficient for the new crop of "mega data centers" that are expected to pop up as the industry consolidates its holdings. Facebook's pod infrastructure is exceptionally scalable, allowing users to simply add pods as their capacity needs increase.
Tech giants have been at the heart of data center design for years. However, a single advancement in data center design has the potential to upset well-established brands with long-standing solutions. That's why IT staffs can minimize their own workload by keeping data center expansion as scalable and flexible as possible. Often makeshift solutions for scalability, such as connecting one network to another site via remote console servers, can act as cost-effective transitional solutions while IT makes decisions about the future of the facility's data architecture.
Perle's wide range of 1 to 48 port Perle Console Servers provide data center managers and network administrators with secure remote management of any device with a serial console port. Plus, they are the only truly fault tolerant Console Servers on the market with the advanced security functionality needed to easily perform secure remote data center management and out-of-band management of IT assets from anywhere in the world.