The complexity of new workloads and increasing demand from consumers means a different server, network and IT needs in today’s hyper scale data center. AT&T, Sprint, and Verizon service providers have strong capabilities of their networks, today many of the large Web-scale providers are now openly revealing their internally developed network designs.
The huge investments done in Software Defined Networking in data centers as well as Wide Area Networks and have published details about their home-grown SDN software and white box switch implementations by Google, Amazon, Facebook, Microsoft, and others. In reality, obtaining high performance, massive scale and low latency is now competition for the hyper-scale cloud providers.
Hyperscale SDN networks differentiate their scale, performance, reliability and their provisioning/management requirements. Web-scale organizations are gigantic in the scale of their compute and storage capacity. Hyperscale data centers generally range in size from 250,000 square feet to more than 500,000 square feet and house 100,000 to more than 500,000 physical servers.
Hyperscale data centers need enormous amounts of network bandwidth between physical and virtual servers, so over the next few years, Ethernet links will be moving from 10GB to 100GB, with potential interval speeds of 25GB, 40GB, and 50GB. Creative thinking is required to Transmit traffic between these huge data centers and/or to the Internet. According to record in Facebook’s new data center network design, for example, a bank of edge switch pods provides 7.68TB/sec of switching capacity up to the Internet or metro DCI. Optical suppliers, like Infinera, Ciena and ALU are moving to deliver dense, cost-effective optical connections to support web scale needs for 100s of 100GB optical links per data center location.
The increasing capacity of the hyper scale data centers makes provisioning network bandwidth an outstanding challenge for the IT organization. Facebook and others have said that their network is too large to physically provision and that automation is a key goal for their SDN implementation.
Web scale providers market their hyper scale SDN networks
Recent Commerce shows such as the Open Network Summit and Open Network User Group, executives from Google, Facebook and Microsoft offered notable details on the scale, performance and architecture of their data center and wide area networks. Their unique network environments have provided incentives to develop their own SDN software and, in the case of Facebook, white box switch hardware. The advantages of these R&D efforts are now starting to trickle down market as the details are publically revealed. The aim of the Web-scale suppliers is to commoditize the network at scale in the same way they have helped to commoditize the server and storage market.
Details on SDN deployments contains:
Google extended SDN principals to create a centralized software control stack to organize thousands of switches within the data center and serve them as one large fabric. Its network allows 100,000 servers to exchange information at 10Gbps within a single data center – a capacity increase of 100x over the last several years.
Facebook’s FBOSS software, which executes a hybrid of distributed and centralized control to manage its network. Facebook launched a Linux-based, top-of-rack network switch called the Wedge that it plans to make available as an open-source hardware design with the help of its Open Compute Project. It’s “6-pack” platform is the core of its new fabric, and it uses “Wedge” as its basic building block.
Microsoft Azure storage and compute utilization is doubling every six months and its cloud platform composed of 22 hyper-scale regions around the world with millions of hosts. The Azure network needs a self-provisioning, virtualized, partitioned and scale-out design, delivered through SDN on commodity servers. Microsoft operates internally designed Vents and Virtual Filtering Platform (VFP) to allow Azure’s data plane to act as a Hyper-V virtual network switch.
Web-scale enterprises have been publicly emphasizing their SDN innovations as a key competitive differentiator. Delivering high performance, massive scale and low latency in both data center and WAN is now competition for many hyper scale cloud providers. These technology proceeds in network design are likely to trickle down to enterprise scale data centers and WANs over the next several years.