Category Archives: Open Switch
Commodity pricing pressure is forcing changes in the top-of-rack switch market. Pluribus Networks, a networking startup that specializes in switches with an operating system that enables network virtualization, is the latest vendor to feel those effects. It will rely on a third-party integrator to manufacture its top-of-rack data center switches from now on. Merchant switch silicon from Broadcom and other chip makers have largely commoditized the top-of-rack switch market. Several startups have emerged with network operating systems designed to run on white box or bare-metal switches. Dell recently became the first OEM switch maker to open up its data center switches to third-party operating systems. Now, Pluribus has decided that selling and supporting software is more important than selling low-margin top-of-rack switches.
Pluribus Networks Brings the Server-Switch Paradigm Into the Broader Commercial Market at White-Box Economics
The Pluribus Freedom Server-Switch architecture developed over the last four years has been validated by the recent Facebook ‘Wedge’ announcement, which combines switching and compute into a modular and open top-of-rack form-factor platform. The Pluribus E-Series based on open off-the-shelf technologies while leveraging the Pluribus NetvisorTM network hypervisor is in fact the first broadly available server-switch in this new category of SDN platforms, unlocking the disruptive potential of new networking business models, while poised to adopt the OCP architecture in the future.
Social media giant Facebook is one step closer to ousting Cisco Systems as its primary datacenter switching supplier with the launch of a new top-of-rack engineering project inside of the company code-named “Wedge.” As in the front of the wedge that is going to split apart the $20 billion and fairly monolithic datacenter switching market. Jay Parikh, vice president of infrastructure engineering at Facebook, divulged some of the details on the homemade Wedge top-of-rack switch at the GigaOm Structure conference in San Francisco.
Years ago, Facebook Inc. announced that it was embarking on an initiative to enable data center operators to use any number of software options after buying a vendor’s hardware. A novel idea, but little has been heard since this long- ago announcement until very recently, when Facebook announced a major development with its Open Compute Project. The Open Compute Project, the name given to this initiative by Facebook, has been successful in moving forward on defining a network switch that could be used on many operating systems to make this a reality.
It has been a year since Facebook announced that its Open Compute Project had an initiative focused on defining a network switch that could be used with a variety of operating systems, so that data center operators would not get locked into using a single vendor’s software once they bought that vendor’s hardware. Facebook’s wish to disaggregate networking hardware from networking software has now been granted. Two switch designs (one by Mellanox and the other by Broadcom) were submitted to Open Compute for approval, and Facebook is already testing a handful of the Broadcom boxes in production in its data centers, Najam Ahmad, director of network engineering at Facebook, said.
The University of Texas at San Antonio is launching one of the largest Open Cloud networks in academia. The network utilizes OpenStack software, co-founded by Rackspace, and Open Compute hardware, founded by Facebook, and will support advanced computing and big data analytics research. The UTSA Cloud and BigData Laboratory, with more than 20 doctoral students, will utilize the network to research new technologies and innovations in various areas of computing. The laboratory was built in collaboration with industry partners such as Rackspace, Open Compute Project Communities, Mellanox, Internet2, ZeroVM and many others.
Mellanox 10 and 40 Gigabit Ethernet NICs Support Microsoft Open Compute Project (OCP) Cloud Server Specification
SUNNYVALE, Calif. & YOKNEAM, Israel–(BUSINESS WIRE)–Mellanox® Technologies, Ltd. (NASDAQ:MLNX), a leading supplier of high-performance, end-to-end interconnect solutions for data center servers and storage systems, today announced 10 and 40GbE NIC support for the Microsoft OCP server and storage specification which are based on hardware used to power Microsoft’s global cloud services, including Windows Azure, Bing, and Office 365. Available now, the ConnectX®-3 Pro OCP-based 10/40GbE NICs with RDMA over Converged Ethernet (RoCE) and overlay network offloads offer optimized application latency and performance while maintaining extremely low system power consumption.
Facebook Inc. said it plans to move this year to new networking equipment in its data centers and stop purchasing gear from conventional networking suppliers. The new gear will be based on standards that are being developed by the Open Compute Project, Jay Parikh, Facebook’s company’s vice president of infrastructure, told CIO Journal. Facebook’s decision could impact incumbent gear providers like Cisco Systems Inc.CSCO, especially if more companies follow suit, but those vendors say the new approaches lack sophistication.
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.
MENLO PARK, Calif. – The Open Compute network switch is moving closer to reality. Broadcom, Intel and Mellanox have each submitted specifications for a top-of-rack switch to the Open Compute Project, and were on hand at Facebook headquarters Monday to promote their contributions to the open-source hardware initiative.
Mellanox Proposes Contribution of 10 Gigabit Ethernet Switch Specification to the Open Compute Project to Enable Cost and Energy-Efficient, Scalable Data Centers
Mellanox® Technologies, Ltd. (NASDAQ:MLNX), a leading supplier of high-performance, end-to-end interconnect solutions for data center servers and storage systems, today announced it is offering its SwitchX®-2 Top-of-Rack switch as a proposed contribution to the Open Compute Project. The SwitchX-2 switch dramatically improves power consumption, latency and density, enabling larger, efficient and more cost-effective data center designs. By open sourcing the designs for the switch, Mellanox is helping drive a faster pace of innovation and a great focus on efficiency and scale in data center technologies.
Just six months ago, we announced our intention to expand the charter of the Open Compute Project to include networking hardware. At the time, it was our hope that we could build on the momentum we’d established in opening up server, storage, and datacenter designs and collaborate with the broader community on the development of an open, OS-agnostic top-of-rack switch. Such a switch, we believed, would enable a faster pace of innovation in the development of networking hardware; 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.
SAN JOSE, Calif. — Broadcom, Intel, and Mellanox have developed competing specifications for datacenter switches, responding to a call from the Facebook-led Open Compute Project. OCP called earlier this year for open specs for software-agnostic leaf and spine switches to complement its existing specs for streamlined servers. The specs aim to speed innovation in networking hardware, “help software-defined networking continue to evolve and flourish,” and give big datacenter operators more flexibility in how they create cloud computing systems, said Frank Frankovsky, a Facebook datacenter executive and chair of the OCP Foundation in a blog posted Monday.
Facebook is a classic example of an exponentially growing business with extreme scale IT needs that cannot do things in the data center the way traditional large enterprises do. It would go broke if it did. Or more precisely, it might have never gotten outside of Mark Zuckerberg’s Harvard dorm room. Facebook’s software and the infrastructure on which it runs are literally the business. The company has been relentless in hacking all parts of its stack, from PHP compilers down to servers and out to data center designs.
Facebook’s infrastructure team has succeeded in getting nearly every part of the data center to do exactly what the applications need it to do – except its network switches. The team has designed its own servers and storage arrays, their data centers have their own mechanical and electrical designs, and even the IT racks have been reinvented.