Facebook and the Open Compute Project (OCP) continued its figurative march through Georgia this week. The next target? The top-of-rack switch, which the OCP intends to replace with its own SDN design built from the ground up.
The OCP team plans to work on the new switch design at the OCP Engineering Summit, which will kick off May 16 at MIT. Frank Frankovsky, the chairman of the OCP and vice president of hardware design and supply chain operations at Facebook, noted in a blog post that, while the Project has managed to build more open hardware designs in the data center, “we’re still connecting them to the outside world using black-box switches that haven’t been designed for deployment at scale and don’t allow consumers to modify or replace the software that runs on them.”
The goal is to engineer an OS-independent solution for the top of rack, he added, with the server itself capable of controlling the switch operation. Najam Ahmad, who runs the network engineering team at Facebook, has volunteered to lead the project.
Controlling the switch via the server is the principle behind Software-Defined Networking (SDN), which means that the OCP is throwing its hat into that popular ring. The May summit at MIT will attract Big Switch Networks, Broadcom, Cumulus Networks, Facebook, Intel, Netronome, VMware, and other companies—all of whom, however ironically, stand to lose if their proprietary products are open-sourced in some way by OCP. Also in attendance at that event will be the OpenDaylight and Open Networking Foundation groups, both of which are more aligned with OCP goals.
In an utterly unsurprising twist, Cisco—whose business will likely be affected most by a vendor-neutral switch—is not among those companies Frankovsky listed as attending.
“It’s our hope that an open, disaggregated switch will enable a faster pace of innovation in the development of networking hardware,” he wrote, and “help software-defined networking continue to evolve and flourish; and ultimately provide consumers of these technologies with the freedom they need to build infrastructures that are flexible, scalable, and efficient across the entire stack.”
Frankovsky concluded by noting that the OCP’s goals were to build disaggregation into the server so that it could “establish common standards that everyone can adopt and build upon, from the bottom of the hardware stack to the top.” The question is, where exactly is the top? And how quickly will the group get there?