Open Compute Project Considers Switch Specs
The Open Compute Project (OCP) is pushing forward with its efforts to develop an open, OS-agnostic top-of-rack switch. Earlier this week, it highlighted four specifications from Broadcom, Cumulus Networks, Intel and Mellanox under consideration by the OCP incubation committee.
The OCP was launched in 2011 by Facebook with a focus on developing open data center components such as racks and servers. The group, now a non-profit foundation, expanded its charter six months ago to include network hardware.
As envisioned by the OCP, an open switch will speed innovation in networking hardware, help SDN evolve, 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," wrote Frank Frankovsky, chairman and president of the OCP Foundation, in a blog post.
The project has attracted some 30 potential contributions covering most of the network hardware stack and some of the network software stack, said Frankovsky, who is also Facebook's VP of hardware design and supply chain operations.
Among the contributions is Broadcom's full specification for and implementation of a 10/40-GbE top-of-rack switch based on the Trident switch architecture. Meanwhile, Intel said its <http://www.intel.com/content/www/us/en/switch-silicon/onp-switch-referen... 10/40-GbE switch specificationwill include all necessary subsystems for switching, control CPU, peripherals, external interfaces, power, cooling and mechanical enclosure.
The third switch specification comes from Mellanox, which proposes contribution of its <http://www.networkcomputing.com/next-gen-network-tech-center/mellanox-ma... SwitchX-2 x86-based top-of-rack switch specification. The company said it would be the first switch to support the Open Network Install Environment (ONIE), developed by Cumulus Networks, on x86.
Cumulus Networks is offering ONIE to OCP to enable installation of any networking OS onto bare-metal switches. According to Cumulus, ONIE will allow companies to use the network operating system of their choice on a variety of hardware.
Rohit Mehra, IDC VP of network infrastructure, said the push for open switches is being driven more by the requirements of cloud service providers than the immediate needs of enterprise IT organizations. Cloud data centers are looking to scale and deploy networks faster to accommodate their growing customer bases. Larger cloud providers also are more likely to have the resources and expertise to mix and match a variety of hardware and software that works best for their environments, and are looking to solve problems quickly, he said.
[To get started with Cisco's software-defined networking platform from Insieme, you'll need Cisco's new Nexus switches. Read the details in "Cisco Unveils Insieme SDN Platform, New Switches."]
Being able to scale is part of the appeal, and there are CAPEX benefits from buying lower cost, industry-standard hardware, such as OEM switches, said Mehra. “But the bigger story is the operational expense. This is where closed versus open comes in," he said.
The open paradigm gives cloud providers more choice, he said, allowing them to cherry pick software to support their needs: Whether it’s orchestration management, security tools, network analytics engines or even homegrown applications, they won’t be constrained by a particular ecosystem.
There are still a lot of unknown implications to Open Compute Project, and plenty of reasons to take advantage of the tried and tested offerings from incumbent switch vendors, such as documentation and vendor support, Mehra said. It will take a few years to clearly to understand how to quickly integrate new functions and also make sure they don’t break existing ones, he said.
“We are going into completely new, uncharted territory.”
Recommended For You
In honor of St. Patrick’s Day, there’s no better time to reflect on those instants when life threw us a curveball, but we were able to hit a home run.
The success of modern enterprises, especially those utilizing real-time communications solutions, is highly reliant on IT infrastructure availability.
To understand the critical role of HTTP/2 in streamlining operations, we must look back at the technologies and implementation gaps that got us where we are today.
A video overview and best practices on how to reduce broadcasts and find other things to tune.
This is a great example of the perfect storm of variables coming together to cause performance issues. Watch the video to see how the problem was found.
Providers should be making infrastructure work for everyone in 2019, improving efficiency and opening up networks for all apps on their infrastructure.