Ethernet Alliance

Ethernet Alliance Blog

How can 400GbE Leverage 100GbE Technology to Lower Cost/bit?

By Scott Kipp

Last week at the IEEE 802.3 Plenary meeting in Orlando, FL, the IEEE 802.3 400 Gb/s Ethernet (400GbE) Study Group was started with unanimous support!  John D’Ambrosia, Dell (also Chair of the Ethernet Alliance), led the Call-For-Interest (CFI) to start the study group.  The 400GbE Study Group should meet for the first time in Victoria, British Columbia in May and the standard should be completed by 2016. 

The IEEE 802.3  400GbE Study Group will define the objectives for the future 400GbE project, and will likely choose from copper, multimode and single-mode fiber implementations.  Ultimately, these objectives will need to be technically feasible and cost effective, while at the same time having broad market potential.   It is thought by many that 400 GbE will be able to re-use existing technology, similarly  to how 40GbE and 100GbE re-used 10GbE technology. 

For example, 400GbE could be defined in such a way that that is uses 100GbE modules to build 400GbE links.  Consider that the industry has already invested millions of dollars in 100GBASE-LR4, which could be  leveraged in defining a single-mode 400GbE specification.  This would allow 400GbE to ride the cost reduction curves of 100GbE technology.

The deployment and interaction between 10GbE and 40GbE in data centers is highlighting some interesting possibilities.  For example, just last week the Ethernet Alliance hosted a panel presentation at OFC on “The Need for Speed – Beyond 100GbE”.   In his presentation, “Stepping Stones to Terabit-Class Ethernet”,  Dr. Jeffery J. Maki, Juniper Networks, suggested a similar concept, where 100GBASE-LR4 specifications  provided the basis for the future 400GbE specification.   The illustration below shows how such a strategy would enable four 100GbE modules to interact as a single 400GbE module.

Blog Pic

 

End users have used parallel multimode fiber (MMF) for years with InfiniBand, inter-chassis links, 40GBASE-SR4, 100GBASE-SR10 and soon 100GBASE-SR4.  This concept could be with single mode fiber as well.  The end-user could run 4 sets of duplex single-mode fiber instead of ribbon fibers as shown.  This concept could reduce the initial deployment of 400GbE by using the existing single mode fibers already deployed and potentially even re-using existing 100GbE modules.  Such a strategy could lower the initial investment in the 400GbE physical layer, which would be a good idea for the industry to consider.

With this technique of using multiple 100GbE modules, a switch or blade could also break out the 400GbE port into 4 100GbE ports.  This breakout would enable high density 100GbE ports in a similar manner to how 40G QSFP+ modules are broken out into 4 10GbE ports today.

The work on 400GbE is just getting started, and ideas such as this and others, will most likely be presented, discussed, and considered.  It will be the role of the IEEE 802.3 400 GbE Study Group to hear these presentations, as it works towards defining the 400GbE project.  Such efforts always drive the need for consensus within the industry, and this is where the Ethernet Alliance will help the industry progress!    

 

By Scott Kipp

President of the Ethernet Alliance

The views and opinions expressed in this blog are solely that of the individual(s) and should not be considered the views or positions of the Ethernet Alliance.

Share

The views and opinions expressed in this blog are solely that of the individual(s) and should not be considered the views or positions of the Ethernet Alliance.

Share

The Ethernet Alliance at OFC 2013 – Demonstrating. Educating. Advancing.

By Matt Traverso

The mission of the Ethernet Alliance is to expand the ecosystem of Ethernet and foster the development of Ethernet technologies.  We have found that one of the best places in the world to showcase the expanding footprint of Ethernet into optical communications is OFC/NFOEC.  This is our third year presenting at OFC, and while I personally have been involved in OFC for over a decade, this is my first year to assist the Ethernet Alliance at OFC. 

We started conversations about OFC2013 nearly a year ago and the discussions were focused on opportunities for interoperability demonstrations with our member companies.  Thanks to the dedication of twelve participating member companies, the Ethernet Alliance is working on a fantastic live demonstration encompassing switches, routers, cable management, as well as discrete components – truly a significant display of the interoperability of Ethernet. 

The live demo showcases multiple data centers interconnected,  via 100GbE connections across routers from three different equipment manufacturers.  Within each data center, various configurations of switches, servers and test equipment will be implemented.  These are connected via a mixture of 10GbE and 40GbE highlighting some of the breakout cables being used to intermix 10GbE and 40GbE in networks today.

Last week the demo participants were helped by the tireless efforts of Andrew Bojak and the rest of the Spirent Communications team, who hosted our “hot stage” of the live demonstration.  On the show floor, we will have live demo of multiple data centers operating across a fully interoperable Ethernet network, including servers, switches, routers, and cabling, and incorporating speeds ranging from 10 to 100GbE. 

On top of the demo preparation, the entire team has been working to prepare our Ethernet Alliance presence at OFC with a theatre built-into our booth, as well as, hosted panels on the broader show floor.  We look forward to promoting Ethernet and interoperability to all of the attendees at OFC 2013. I invite all of you to stop by and visit us at booth #2909.

Matt Traverso is an Ethernet Alliance board member and the hardware engineering manager at Cisco Systems.

Share

The views and opinions expressed in this blog are solely that of the individual(s) and should not be considered the views or positions of the Ethernet Alliance.

Share

New Ethernet Alliance TeraFabric Plugfest Technical White Paper Shows the Future is in Sight

By Patrick Strick

The Ethernet Alliance often holds plugfests where vendors working within specific technology areas like 10GbE or Data Center Bridging come together for weeklong equipment interoperability testing, and checking and verifying compliance with respective specifications.

In 2012, we took this concept to an entirely new level by holding a broad-focus event – the TeraFabric Plugfest – encompassing multiple technologies within the Ethernet ecosystem. Among the diverse technologies represented included 40Gb Ethernet, the IEEE-ratified version of data center bridging (802.1Qaz), 10 and 40Gb active optical cables, QSFP 40Gb modular connectors (40GBASE-CR4), 40Gb to 4x10Gb fan-out cables, and 10Gb over twisted-pair cabling (10GBASE-T). Among the protocols tested were Fibre Channel over Ethernet (FCoE) and iSCSI DCB running multiple priorities and multiple lossless classes…and then we put it all together by simulating a real-world data center environment capable of 1 terabit of throughput. Not bad for just a week’s time.

Accomplishing this required bringing together some 20 vendors from across the Ethernet industry with their equipment at the University of New Hampshire’s InterOperability Laboratory – no small feat in itself. But we managed to do so, and at the end of October 2012, the TeraFabric Plugfest testing and interoperability event was launched. Turnout for TeraFabric was excellent, with participants representing a diverse cross-section of the Ethernet landscape. Companies taking part included Amphenol Corporation, Applied Micro, Broadcom Corporation, Cisco Systems, CommScope, Inc., Dell Networking and Dell EqualLogic, Emulex Corporation, Foxconn International Holdings Ltd., Intel Corporation, Ixia, JDS Uniphase Corp, Leviton Manufacturing Co., NetApp Inc., Nexans S.A., Panduit Corp., PLX Technology, Inc., QLogic Corporation, Samtec, Inc., Siemon Company, TE Connectivity Ltd., and Volex.

The TeraFabric Plugfest was an undeniably valuable event for our member companies, and we want to share what we have learned with the broader community, as well. To do so, the Ethernet Alliance is today publishing an informative white paper describing the approach taken and testing accomplished during this first-of-its-kind plugfest. The takeaway for data center managers and the Ethernet industry as a whole is this: Not only is the next generation of converged Ethernet networks here – the technology is sound, truly interoperable, and ready for real-world deployments and broad-scale adoption – the generation beyond that is already in sight.

You can learn more by downloading your copy of about the Ethernet Alliance’s TeraFabric Plugfest white paper now: http://bit.ly/ZcMpFj.

Patrick Strick

Technical Chair, Data Center Subcommittee, Ethernet Alliance

Technical Marketing Engineer, Data Center Platforms, NetApp

Share

The views and opinions expressed in this blog are solely that of the individual(s) and should not be considered the views or positions of the Ethernet Alliance.

Share