Optical components enter an era of technology-pull
Gazettabyte asked ADVA Optical Networking, Ciena, Cisco Systems and Ovum about their impressions following the recent OFC/NFOEC 2012 exhibition and conference.
OFC/NFOEC reflections: Part 2
"As the economy continues to navigate its way through yet another very difficult period, it was good to see so many companies innovating and introducing solutions."
Massimo Prati, Cisco Systems
Massimo Prati, Cisco Systems
For Cisco Systems, 100 Gigabit was a key focus at the show. "There were many system and component vendors, including Cisco, demonstrating newly available, economically feasible 100 Gig innovations," says Massimo Prati, vice president and general manager for Cisco.
Linking data centres was another conference theme. "Inter-data centre connectivity continues to focus on scalable and simple solutions in long-haul and metro networks connecting data centres worldwide." Cisco believes metro 100 Gigabit deployments will become prevalent in 2013 and 2014, especially if low‐cost coherent technology becomes available.
"A dedicated workshop focused on data centre architectures, held on the first day of the conference, was heavily attended," says Prati. "So certainly the link between cloud and optical is being established and is a key driver for high-speed transport networks."
Another conference theme was interconnect within the data centre, and the need for photonic integration for low‐cost, low‐power links, says Prati: "From a Cisco standpoint, several of our customers were pleasantly surprised by our recently completed acquisition of Lightwire, which develops advanced optical interconnect technology for high-speed networking applications." Lightwire is a silicon photonics startup that Cisco acquired recently for US $271 million.
What Cisco says it learned from OFC/ NFOEC was that service providers are planning 100Gbps deployments within the next 12 months and are looking at second- and third-generation solutions. "There is quite a bit of energy around future upgrades to 400 Gig and one Terabit transport solutions, but service providers continue to monitor if and how these solutions will operate within their existing fibre plants."
Prati expects more industry consolidation. "With the influx of 100 Gig solutions, it appears we may be ripe for further consolidation within the industry, particularly further down the technology food chain," he says.
He also remains optimistic about the industry's prospects.
"We believe that the excitement around high-speed, long-haul transport, combined with cloud and data centre innovation, continues to fuel a lot of new product solutions and architectures," he says. "Content providers like Google and Facebook have clearly expressed interest in optical technologies addressing their issues with bandwidth demands and need for high-speed interconnect for their data centres."
Joe Berthold, Ciena
Whereas last year there was much discussion about of the next rate for Ethernet - 400 Gig or one Terabit - this year 400 Gigabit had most mindshare, says Joe Berthold, vice president of network architecture at Ciena. "I barely heard any mention of one Terabit in the context of a contest with 400 Gigabit," he says.
"I could hear some rumblings about alternative form factors – which might lead to fragmentation of the market"
Joe Berthold, Ciena
400 Gigabit was given a boost with the line-side transmission component announcements. Ciena announced its WaveLogic3 and Alcatel-Lucent detailed its Photonic Service Engine.
Another noteworthy development was the buzz around silicon photonics, stirred in part by Cisco's Lightwire acquisition. "Silicon photonics has passed from a technology of research interest to one that has progressed to serious development," says Berthold. "Data centre interconnects look like a promising initial application."
There was no developments at the show that surprised Berthold. But he is concerned about the potential for proliferation of 100 Gigabit client-side form factors, especially for pluggable modules.
"I am going under the assumption that there is still broad industry support for the CFP progression - from the current CFP to a CFP2 followed by a CFP4 for single-mode fiber applications over metro distances," he says.
Even though there are a variety of technologies appearing in the CFP form factor, this common physical module has helped control system development cost. "I could hear some rumblings about alternative form factors – which might lead to fragmentation of the market," he says.
Berthold is encouraged by the broad base of development efforts underway, particularly for 100Gbps transceivers, but also lower-cost 10Gbps and 40Gbps client-side modules. He notes the progress in reducing the cost of 100 Gigabit client interfaces over the next year. "Their high cost has held back adoption of 100 Gig," says Berthold. "We have had very cost effective 10 Gig multiplexing technology to fall back on, but it looks like native 100G interfaces are poised for growth."
Jörg-Peter Elbers, ADVA Optical Networking
Jörg-Peter Elbers, vice president, advanced technology at ADVA Optical Networking, was struck by the wide range of hot topics discussed at the show.
These include software-defined optics based on programmable transceivers that use advanced DSP technology and flexgrid ROADMs as the basis of a new coherent express layer. He also notes that control plane technologies are becoming an essential asset in managing network complexity when unleashing untapped network capacity.
"Traffic and content keeps growing at exponential scale - the fundamental demand-drivers are intact"
Jörg-Peter Elbers, ADVA Optical Networking
Meanwhile, the rapid increase in end-user traffic, specifically mobile, is driving PON. As a result WDM is moving closer to the network edge, entering aggregation and access networks. He believes dense WDM-PON is gaining traction for mobile backhaul as fibre becomes the bottleneck when moving from Long Term Evolution (LTE) to the LTE-Advanced cellular technology.
Other trends to note, he says, are software-defined networking (SDN) and OpenFlow. "Originating from the campus and data centre world, network programmability is increasingly seen as key for tighter integration, more automation, and virtualisation of IT and computing services," says Elbers.
The industry increasingly sees the metro market as important to ramp up 100Gbps volumes, with different modulation solutions being promoted by vendors. These include performance reduced 100Gbps DP-QPSK (dual polarisation, quadrature phase-shift keying), 200Gbps DP-16QAM (dual polarisation, 16-quadrature amplitude modulation) and 4x28G direct-detection.
While some people expressed concerns about a fragmentation of the 100 Gig market, power consumption, footprint and cost are of primary importance in the metro, he says. "One analyst at the Ovum 100Gbps metro workshop at OFC said: 'Maybe, for a hammer everything looks like a nail…'," says Elbers. "With 4x28G optical duobinary being able to make use of 10Gbps T-XFP/SFP+, IEEE 802.3ba and CFP technologies, we believe there is a justification to differentiate."
ADVA demonstrated its 4x28Gbps optical duobinary direct-detection product at the show.
Elbers noted an interest in multi-core and few-mode fibres. "The next x10 in bandwidth is difficult to reach as additional gains from amplification, modulation, FEC and denser carrier spacing will be limited." he says. "The research community therefore is looking into new fibre types to add the spatial and modal dimensions alongside the current optimisation strategy." An area interesting to watch, but fundamental technical and economic challenges remain, he says.
He too is optimistic about the industry's prospects: "Traffic and content keeps growing at exponential scale - the fundamental demand drivers are intact." As a result, optical innovation will play an even bigger role in the future to keep pace with the bandwidth growth, he says.
Karen Liu, Ovum
"We're clearly in a technology-pull phase rather than technology-push phase with multiple system vendors doing 400Gbps-capable stuff instead of component guys showing demonstrations years in advance of system activity," says Karen Liu, principal analyst, components telecoms at Ovum.
"Optical burst mode switching may be crossing over from rather 'pie-in-the-sky' to practical"
Karen Liu, Ovum
It is not that that the components vendors aren't making innovative products, she says, just that they are not making announcements until there is real demand. "Corning, for example, showed a fiber that has already been shipping into Lightpeak," says Liu.
What surprised Liu at the show was Huawei's optical burst transport network prototype. "Optical burst mode switching may be crossing over from rather 'pie-in-the-sky' to practical," says Liu.
She notes how there isn't as much optics-versus-electronics positioning anymore but more a case of optics working with electronics. "Huawei's OBTN is an example," says Liu. "Instead of using optical burst mode to make an all-optical network, optics is part of a hybrid design."
Liu says there are now multiple relationships between silicon and optics including the two working together instead of in competition. "In networking, the term translucent networks seems to have gained popularity."
Reflections 2011, Predictions 2012 - Part 2
Gazettabyte asked industry analysts, CEOs, executives and commentators to reflect on the last year and comment on developments they most anticipate for 2012. Here are the views of Verizon's Glenn Wellbrock, Professor Rod Tucker, Ciena's Joe Berthold, Opnext's Jon Anderson, NeoPhotonics' Tim Jenks and Vladimir Kozlov of LightCounting.
Glenn Wellbrock, Verizon's director of optical transport network architecture & design
The most significant accomplishment from an optical transport perspective for me was the introduction of 100 Gigabit into Verizon's domestic - US - network.
"The key technology enabler in 2012 will be the flexible grid optical switching that can support data rates beyond 100 Gigabit"
That accomplishment has paved the way for us to hit the ground running in 2012 with a very aggressive 100 Gigabit deployment plan. I also believe this accomplishment gives others the confidence to start taking advantage of this leading-edge technology.
With coherent receiver technology and the associated high-speed electronics lowering the propagation latency by up to 15%, we see a much cleaner line system design that eliminates external dispersion compensation fibre while bringing down the cost, space and power per bit.
The value of the whole industry moving in this direction means higher volumes and, therefore, lower costs. This new infrastructure will allow operators to get ahead of customer demand, thus improving delivery intervals and introducing new, higher bandwidth services to those large key customers that require it.
In my opinion, the key technology enabler in 2012 will be the flexible grid optical switching that can support data rates beyond 100 Gigabit and provides the framework to support colourless, directionless and contentionless optical nodes.
Today, field technicians must plug a new transmitter/ receiver into the appropriate direction and filter port at both circuit ends. With this new technology, operations personnel can simply plug the new card into the next available port and it can then be provisioned, tested and even moved to a new colour or direction remotely without any on-site personnel involvement - even when there are multiple copies of the same colour on the same add/ drop structure coming from different fibres.
This new nodal architecture takes advantage of the inherent channel selection capability of the coherent receiver to eliminate fixed filters and opens up the door for a truly reconfigurable optical add/ drop multiplexer (ROADM) - creating new flexibility that can be used for optical restoration, network defragmentation, operational simplicity, and more.
Rod Tucker, Director of the Institute for a Broadband Enabled Society (IBES), Director of the Centre for Energy-Efficient Telecommunications (CEET), and professor of electrical and electronic engineering at the University of Melbourne.
Australia's National Broadband Network (NBN) hit the ground running in 2011.
The project is still many years from completion, but in 2011 the roll-out of fibre-to-the-premises infrastructure began in earnest. This is a very noteworthy project - a wholesale broadband access network delivering advanced broadband services to the entire population of the country, including fibre to 93% of all premises and a mixture of fixed wireless and satellite to the remainder. At an estimated cost of around AUS$36 billion, the price tag is not small.
"The environment created by [Australia's] National Broadband Network will greatly enhance opportunities for innovations in new services and new modes of broadband service delivery"
But the wholesale-only model maximises opportunities for competition at the service provider level, and reduces wasteful duplication of infrastructure in the last mile. A remarkable aspect of the NBN project is that a deal has been struck between the incumbent telco, Telstra, and the government-owned owner of the NBN.
Under this deal, Telstra will shut down its Hybrid-Fibre-Coax (HFC) network and decommission its legacy copper access network. Australia will become a truly fibre-connected country, with a future-proof broadband infrastructure.
My thoughts for 2012 also relate to Australia's National Broadband Network. The environment created by the NBN will greatly enhance opportunities for innovations in new services and new modes of broadband service delivery.
I anticipate that in 2012 and beyond, new services providers and aggregators in areas such as health care, education, entertainment and energy will emerge.
I am very excited about the opportunities.
Joe Berthold, vice president of network architecture at Ciena
One of the most memorable developments from a network architecture point of view was the clear emergence of the category of packet-optical switching products to serve as the transport layer of backbone IP networks.
For years two competing points of view have been put forth. First, in the 'IP-over-glass' position, long-haul optics is incorporated into core routers. This has never taken off, with some disappointing attempts in the early days of 40 Gigabit. The second approach involves a separate, very much simpler, packet optical transport platform being introduced to interconnect core routers. The packet transport could be based on Ethernet protocols, MPLS, MPLS-TE or MPLS-TP.
"It will be interesting to see if a large internet data centre operator decides to embrace the OpenFlow concept at this very early stage of its development"
What is quite significant in this development, traditional router vendors seem to be going in this direction too, with the vision of a much simpler packet switching platform to keep cost, space and power under control.
This is a clear response to the overwhelming need we see in the market, representing a separation of packet switching into two layers: one with global routing capability at strategic locations in the network, and the other with flexible transport functionality for network traffic engineering.
In 2012 it will be fascinating to see how the struggle for protocol dominance plays out within the data centre.
While the IETF has many competing proposals, worked in multiple groups, the IEEE is in final ballot now for Shortest Path Bridging (IEEE 802.1aq).
Shortest Path Bridging has broad applicability in networks, but we might see it first emerge as a solution within the data centre.
The other contender within the data centre is OpenFlow, which has developed quite a momentum too.
It will be interesting to see if a large internet data centre operator decides to embrace the OpenFlow concept at this very early stage of its development.
Jon Anderson, director of technology programme at Opnext
Our most significant 2011 events were the Japan great earthquake in March and the Thailand floods in October. Both events caused major disruptions and challenges in optical component supply-chain management and manufacturing.
JDS Uniphase's tunable SFP+ announcement was well ahead of the technology curve.
"Our most significant 2011 events were the Japan great earthquake in March and the Thailand floods in October."
In 2012 we expect initial production shipments and deployment of 100Gbps PM-QPSK/ coherent modules. Also a fast production ramp of 40 Gigabit Ethernet (GbE) QSFP+ modules for data centre applications.
Another development to watch is the next-generation 100 GbE interconnect technology and standards development for low-cost, high-density modules for data centre applications.
Lastly, there will be an increased focus on technologies and solutions for 100 Gigabit DWDM in metro and extended reach enterprise applications.
Tim Jenks, CEO of NeoPhotonics
NeoPhotonics made significant progress this year in developments of components and technologies for coherent transmission networks, including receivers, transmitters and advanced approaches toward switching.
We continue to see increasing adoption of coherent transmission systems, broad-scale deployment of access networks and a continuing emergence of large scale data centres as a prominent element of the communications network landscape.
Vladimir Kozlov, CEO of LightCounting
The industry was strong enough to get over an earthquake, tsunami and flood in 2011. Softer demand for optics in 2011 helped - is still helping - many vendors to ride the disruptions. Ironically, the industry was more stressed ramping up production in 2010 to meet demand than dealing with the disruptions of 2011. We are looking forward to a smoother ride in 2012, as demand/ supply reach equilibrium and nature cooperates.
"Ironically, the industry was more stressed ramping up production in 2010 to meet demand than dealing with the disruptions of 2011"
Service provider revenue and capex were up significantly in 2011. Mobile data is driving the growth, but even wireline revenues are improving and FTTx is probably behind it. This should be a sustainable trend for 2012-2015, even as service providers curb expenses to improve profitability, a larger fraction of capex will be spend on equipment. New technology is critical to stay ahead of competition.
Data centre optics had another good year with 10GBASE-T falling further behind schedule and with 100 Gigabit generating much action. This will probably get even more interesting in 2012.
Our conservative forecast for active optical cable, criticised by some vendors, was not conservative enough in 2011. It will take a while for this segment to unfold.
For Part 1, click here
For Part 3, click here
Fulcrum's Alta switch chips add programmable pipeline to keep pace with standards
Part 2: Ethernet switch chips
Fulcrum Microsystems has announced its latest FocalPoint chip family of Ethernet switches. The Alta FM6000 series family supports up to 72 10-Gigabit ports and can process over one billion packets a second.
“Instead of every top-of-rack switch having a CPU subsystem, you could put all the horsepower into a set of server blades”
Gary Lee, Fulcrum Microsystems
The company’s Alta FM6000 series is its third generation of FocalPoint Ethernet switches. Based on a 65nm CMOS process, the Alta switch architecture includes a programmable packet-processing pipeline that can support emerging standards for data centre networking. These include Data Center Bridging (DCB), Transparent Interconnection of Lots of Links (TRILL), and two server virtualisation protocols: the IEEE 802.1Qbg Edge Virtual Bridging and the IEEE 802.1Qbh Bridge Port Extension.
Why is this important?
Data centre networking is undergoing a period of upheaval due to server virtualisation. Data centre operators must cope with the changing nature of traffic flows, as the predominant traffic becomes server-to-server (east-west traffic) rather than between the servers and end users (north-south).
In turn, IT staff want to consolidate the multiple networks they must manage - for LAN, storage and high-performance computing - onto a single network based on DCB.
They also want to reduce the number of switch platforms they must manage. This is leading switch vendors to develop larger, flatter architectures; instead of the traditional three tiers of switches, vendors are developing sleeker two-tier and even a single-layer, logical switch architecture that spans the data centre.
“There are people out there that have enterprise gear where their data centre connection has to go through the access, aggregation and core [switches],” says Gary Lee, director of product marketing at Fulcrum Microsystems. “They may not want to swap out that gear so they are going to continue to have three tiers even if it is not that efficient.”
But other customers such as large cloud computing players do not require such a switch hierarchy and its associated software complexity, says Lee: “They are the ones that are driving to a ‘lean core’, made up of top-of-rack and the end-of-row switch that acts as a core switch.”
Switch vendor Voltaire, a customer of Fulcrum’s ICs, uses such an arrangement to create 288 10-Gigabit Ethernet ports based on two tiers of 24-port switch chips. With the latest 72-port FM6000 series, a two-tier architecture with over 2,500 10-Gigabit ports becomes possible. “The software can treat the entire structure of chips as a single large virtual switch,” says Lee.
Alta architecture
Fulcrum's Alta FM6000 series architecture. Source: Fulcrum MicrosystemsFulcrum’s FocalPoint 6000 series comprises nine devices with capacities from 160 to 720 Gigabit-per-second (Gbps). The Alta chip architecture has three main components:
- Input-output ports
- RapidArray shared memory and
- the FlexPipe array pipeline.
Like Fulcrum’s second generation Bali architecture, the 6000 series has 96 serial/ deserialiser (serdes) ports but these have been upgraded from 3.125Gbps to 10Gbps.“We have very flexible port logic,” says Lee. “We can group four serdes to create a XAUI [10 Gigabit Ethernet] port or create an IEEE 40 Gigabit Ethernet port.”
RapidArray is a single shared memory which can be written to and read from at full line rate from all the ports simultaneous, says Fulcrum. Each memory output port has a set of eight class-of-service queues, while the shared memory can be partitioned to separate storage traffic from data traffic.
“The shared memory design is where we get the low latency, and good multicast performance which people in the broadband access market like for video distribution,” says Lee.
The architecture’s third main functional block is the FlexPipe array pipeline. The pipeline, new to Alta, is what enables up to a billion 64 byte packets to be processed each second. The packet-processing pipeline combines look-up tables and microcode-programmable functional blocks that process a packet’s fields. Being able to program the array pipeline means the device can accommodate standards’ changes as they evolve, as well as switch vendors’ proprietary protocols.
OpenFlow
The FocalPoint software development kit that comes with the chips supports OpenFlow. OpenFlow is an academic initiative that allows networking protocols to be explored using existing hardware but it is of growing interest to data centre operators.
“It creates an industry-standard application programming interface (API) to the switches,” explains Lee. It would allow the likes of a Google or a Yahoo! to switch vendors’ switch platforms as long as both vendors supported OpenFlow.
OpenFlow also establishes the idea of a central controller that would run on a server to configure the network. “Instead of every top-of-rack switch having a CPU subsystem, you could put all the horsepower into a set of server blades,” says Lee. This promises to lower the cost of switches and more importantly enable operators to unshackle themselves from switch vendors’ software.
Lee points out that OpenFlow is still in its infancy. But Fulcrum has added an ‘OpenFlow vSwitch stub’ to its software that translates between OpenFlow APIs and FocalPoint APIs.
What next?
Fulcrum says it continues to monitor the various evolving standards such as DCB, TRILL and the virtualisation work. Fulcrum is also getting requests to support latency measurement on its chips using techniques such as synchronous Ethernet to ensure service level agreements are met.
As for future FocalPoint designs, these will have greater throughput, with larger table sizes, packet buffers and higher-speed 100 Gigabit Ethernet interfaces.
Meanwhile all nine FM6000 series’ chip members will be available from the second quarter, 2011.
Click here for Part 1: Single-layer switch architectures
Click here for Part 3: Networking developments