Do multi-source agreements benefit the optical industry?
System vendors may adore optical transceivers but there is a concern about how multi-source agreements originate.
Optical transceiver form factors, defined through multi-source agreements (MSAs), benefit equipment vendors by ensuring there are several suppliers to choose from. No longer must a system vendor develop its own or be locked in with a supplier.
“Personally, the MSA is the worst thing that has happened to the optical industry”
Marek Tlaka, Luxtera
Pluggables also decouple optics from the line card. A line card can address several applications simply by replacing the module. In contrast, with fixed optics the investment is tied to the line card. A system can also be upgraded by swapping the module with an enhanced specification version once it is available.
But given the variety of modules that datacom and telecom system vendors must support, there are those that argue the MSA process should be streamlined to benefit the industry.
Traditionally, several transceiver vendors collaborate before announcing an MSA. The CFP MSA announced in March 2009, for example, was defined by Finisar, Opnext and Sumitomo Electric Device Innovations. Since then Avago Technologies has become a member.
“The industry has an interesting model,” says Niall Robinson, vice president of product marketing at Mintera. “A couple of companies can get together, work behind closed doors and announce suddenly an MSA and try to make it defacto in the market.”
Robinson contrasts the MSA process with the Optical Interconnecting Forum’s (OIF) 100Gbps line side work that defined guidelines for integrated transmitter and receiver modules. Here service providers and system vendors also contributed. “It was a much more effective and fair process, allowing for industry collaboration,” says Robinson
Matt Traverso, senior manager, technical marketing at Opnext, and involved in the CFP MSA, also favours an open process. “But the view that the way MSAs are run is not open is a bit of a fallacy,” he says.
“Any MSA that is well run requires iteration with suppliers,” says Traverso. The opposite is also true: poorly run MSAs have short lives, he says. Having too open a forum also runs the risk of creating a one-size-fits-all: “One vendor may want to use the MSA as a copper interface while a carrier will want it for long-haul dense WDM.”
Optical transceiver vendors benefit in another way if they are the ones developing MSAs. “Transceiver vendors will not make life tough for themselves,” says Padraig OMathuna, product marketing director at optical device maker, GigOptix. “If MSAs are defined by system vendors, [transceiver] designs would be a lot more challenging.”
Avago Technologies argues for standards bodies to play a role especially as industry resources become more thinly spread.
“MSAs are not standards; there are items left unwritten and not enough double checking is done,” says Sami Nassar, director of marketing, fiber optic products division at Avago Technologies. There are always holes in the specifications, requiring patches and fixes. “If they [transceivers] were driven by standards bodies that would be better,” says Nassar.
Organisations such as the IEEE don’t address packaging and connectors as part of their standards work. But this may have to change. “The real challenge, as the industry thins out, is ensuring the [MSA] work is thorough,” says Dan Rausch, Avago’s senior technical marketing manager, fiber optic products division. “The challenge for the industry going forward is ensuring good engineering and more robust solutions.”
Marek Tlalka, vice president of marketing at Luxtera, goes further, questioning the very merits of the MSA: “Personally, the MSA is the worst thing that has happened to the optical industry.”
Unlike the semiconductor industry where a framer chip once on a line card delivers revenue for years, a transceiver company may design the best product yet six months later be replaced by a cheaper competitor. “The return on investment is lost; all that work for nothing,” says Tlalka.
“Is it a good development or not? MSAs are out there,” says Vladimir Kozlov, CEO of optical transceiver market research firm, LightCounting. “It helps system vendors, giving them a freedom to buy.”
But MSAs have squeezed transceiver makers, says Kozlov, and he worries that it is hindering innovation as companies cut costs to maximize their return on investment.
“There is continual pressure to reduce the price of optics,” adds Daryl Inniss, Ovum’s practice leader components. If operators are to provide video and high definition TV services and grow revenues then bandwidth needs to become dirt cheap. “Even today optics is not cheap,” says Inniss. Certainly MSAs play an important role in reducing costs.
“The transceiver vendors’ challenge is our benefit,” admits Oren Marmur, vice president, optical networking line of business, network solutions division at system vendor, ECI Telecom. “But we have our own challenges at the system level.”
Photonic integration: Bent on disruption
“This is a general rule: what starts as a series of parts loosely strung together, if used heavily enough, congeals into a self-contained unit.”
W. Brian Arthur, The Nature of Technology
Infinera's Dave Welch: PICs are fibre-optic's current disruption
Dave Welch likes to draw a parallel with digital photography when discussing the use of photonic integration for optical networking. “The CMOS photodiode array – a photonic integrated circuit - changed the entire supply chain of photography,” says Welch, the chief strategy officer at Infinera.
Applied to networking, the photonic integrated circuit (PIC) is similar, argues Welch. It benefits system cost by integrating individual optical components but it delivers more. “All the value – inherently harder to pin down - of networking efficiency of a system that isn’t transponder-based,” says Welch.
Just how disruptive a technology the PIC proves to be is unclear but there is no doubting the growing role of optical integration.
“Integration is a key part of our thinking,” says Sam Bucci, vice president, WDM, Alcatel-Lucent’s optics activities. When designing a new platform, Alcatel-Lucent surveys components and techniques to identify disruptive technologies. Even if it chooses to implement functions using discrete components, the system is designed taking into account future integrated implementations.
“We are seeing interest [in photonic integration] across the spectrum," says Stefan Rochus, vice president of marketing and business development at CyOptics. "Long-haul, metro, access and chip-to-chip - everything is in play."
The drivers for optical integration’s greater use are harder to pin down.
Operators must contend with yearly data traffic growth estimated at between 45 to 65 percent yet their revenues are growing modestly. “It’s no secret that the capacity curve - whether the line side or the client side - is growing at an astonishing clip,” says Bucci.
The onus is thus on equipment and component makers to deliver platforms that reduce the transport cost per bit."Delivering more for less," says Graeme Maxwell, vice president of hybrid integration at CIP Technologies. “Space is a premium, power is an issue, operators want performance maintained or improved – all are driving integration.”
Cost is an issue for optical components with yearly price drops of 20 percent being common. “Hitting the cost-curve, we have run out of ways to do that with classic optics,” says Sinclair Vass, commercial director, EMEA at JDS Uniphase.
High-speed optical transmission at 40 and 100 Gigabit per second (Gbps) requires photonic integration though here the issues are as much performance as cost reduction. Indeed, its use can be viewed as the result of the integration between electronics and optics. To address optical signal impairments, chips must work at the edge of their performance, requiring the optical signal to be split into slower, parallel streams. Such an arrangement is ripe for photonic integration.
“It is as if there are two kinds of integration: at the boundary between optics and electronics, and the purely optical planar waveguide stuff,” says Karen Liu, vice president, components and video technologies at market research firm Ovum.
The other market where the full arsenal of optical integration techniques – hybrid and monolithic integration – is being applied is optical transceivers for passive optical networking (PON). Here the sole story is cost.
As old as the integrated circuit
Photonic integration is not new. The idea was first mooted in a 1969 AT&T Bell Labs’ paper that described how multiple miniature optical components could be interconnected via optical waveguides made using thin-film dielectric materials. But so far industry adoption for optical networking has been limited.
"Two kinds of integration: at the boundary between optics and electronics, and the purely optical planar waveguide stuff"
Karen Liu, Ovum
Heavy Reading, in a 2008 report, highlighted the limited progress made in photonic integration in recent years, with the exception of Infinera, a maker of systems based on a dense wavelength division multiplexing (DWDM) 10x10Gbps monolithically integrated PIC.
“Infinera has held very consistently to its original story, including sub-wavelength grooming, and have made progress over time,” says Liu. But she points out a real disruptive impact has not yet been seen: “The problem with the digital camera analogy is that something that is disruptive is not a straight replacement but implies the next step: changing the network architecture, not just how a system is implemented.”
On-off keying to phase modulation
One way operators are accommodating traffic growth is cramming more data down a fibre. It is this trend- from 10Gbps to 40 and 100Gbps lightpaths - that is spurring photonic integration.
“If you look at current 100 Gigabit, it is a bigger configuration than we would like,” says Joe Berthold, Ciena’s vice president of network architecture. Ciena’s first 100Gbps design requires three line cards, taking five inches of rack space, while its second-generation design will fit on a single, two-inch card.
Both 40 and 100Gbps transmissions must also operate over existing networks, matching the optical link performance of 10Gbps despite dispersion being more acute at higher line speeds. To meet the challenge, the industry has changed how it modulates data onto light. Whereas previous speed increments up to 10Gbps used simple on-off keying, 40 and 100Gbps use advanced modulation schemes based on phase, or phase combined with polarisation.
The modulation schemes split the optical signal into parallel paths to lower symbol rates. For example, 40Gbps differential quadrature phase-shift keying (DQPSK) uses two signals that effectively operating at 20 Gigabaud. Halving the rate relaxes the high-speed electronics requirements at the expense of complicating the optical circuitry.
The concept is extended further at 100Gbps. Here polarisation is combined with phase modulation (either DQPSK, or QPSK if coherent detection is used) such that four signals are used in parallel, each operating at 28 Gigabaud.
“The 40/100G area is shaping up to be the equivalent of breaking the sound barrier”
Brad Smith, LightCounting
“Optical integration is becoming a necessity because of 40 and 100 Gigabit [transmission],” says Berthold. “The modulation formats require you to deal with signals in parallel, and using non-integrated components explodes the complexity.”
The Optical Internetworking Forum organisation has chosen dual-polarisation QPSK (DP- QPSK) as the favoured modulation scheme for 100Gbps and has provided integrated transmitter and receiver module guidelines to encourage industry convergence on common components. In contrast, for 40Gbps several designs have evolved: differential phase-shift keying (DPSK) through to DQPSK and DP-QPSK.
Kim Roberts, Nortel's director of optics research, while acknowledging that optical integration benefits system footprint and cost, downplays its overall significance.
For him, the adoption of coherent systems for 40 and 100Gbps – Nortel was first to market with a 40Gbps DP-QPSK system – move the complexity ‘into CMOS’, leaving optics to perform the basic functions. “I don’t see an overwhelming argument for integration,” says Roberts. “It’s useful and shows up in lower cost and smaller designs but it’s not a revolution.”
Meanwhile, optical component companies are responding by integrating various building blocks to address the bulkier 40 and 100Gbps designs.
NeoPhotonics is now shipping two PICs for 40 and 100Gbps receivers: a DQPSK demodulator based on two delay-line interferometers (DLIs) and a coherent mixer for a DP-QPSK receiver.
The DLI, as implied by the name, delays one of the received symbols and compares it with the adjacent received one to uncover the phase-encoded information. This is then fed to a balanced detector - a photo-detector pair. For DQPSK, either two DLIs or a single DLI plus 90-degree hybrid are required along with two balanced receivers.
For 100Gbps, a DP-QPSK receiver has four channels - a polarisation beam splitter separates the two polarisations and each component is mixed with a component from a local reference signal using a 90-degree hybrid mixer. The two hybrid mixers decompose the referenced phase to intensity outputs representing the orthogonal phase components of the signal and the four differential outputs are fed into the four balanced detectors (Click here for OIF document and see Fig 5).
Neophotonics’ coherent mixer integrates monolithically all the demodulation functions between the polarisation beam splitter and the balanced photo-detectors.
The company has both indium phosphide and planar lightwave circuit (PLC) technology integration expertise but chose to implement its designs using PLC technology. “Indium phosphide is good for actives but is not good for passives and it is very expensive,” says Ferris Lipscomb, vice president of marketing at NeoPhotonics.
One benefit of using PLC for demodulation is that the signal path lengths need to have sub-millimeter accuracy to recover phase; implementing a discrete design using fibre to achieve such accuracy is clearly cumbersome.
Another development that reduces size and cost involves the teaming of u2t Photonics, a high-speed photo-detector and indium phosphide specialist, with Optoplex and Kylia, free-space optics DLI suppliers. The result is a compact DPSK receiver that combines the DLI and balanced receiver within one package. Such integration at the package level reduces the size since fibre routeing between separate DLI and detector packages is no longer needed. The receiver also cuts cost by a quarter, says u2t.
Jens Fiedler, vice president sales and marketing at u2t Photonics, acknowledges that the free-space DLI design may not be the most compact design but was chosen based on the status of the various technologies. “We needed to provide a solution and PLC was not ready,” he says.
u2t Photonics is investigating a waveguide-based DLI solution and is considering indium phosphide and, intriguingly, gallium arsenide. “Indium phosphide has the benefit of integrating the DLI with the balanced detector,” says Fiedler. “There are benefits but also technical challenges [with indium phosphide].”
At ECOC 2009 in September, u2t announced a multi-source agreement (MSA) with another detector specialist, Picometrix, which supports the OIF’s DP-QPSK coherent receiver design. The MSA defines the form factor, pin functions and locations, and functionality of the receiver package holding the balanced detectors, targeted at transponder and line card designs.
Photonic integration for high-speed transmissions is not confined to the receiver. Oclaro has developed a 40Gbps DQPSK monolithic modulator. Implemented in indium phosphide, the modulator could even be monolithically integrated with the laser but Oclaro has said that there are performance benefits such as signal strength in keeping the two separate.
Infinera, meanwhile, eschews transponders in favour of its 100Gbp indium phosphide-based PIC.
Take your PIC
In September it announced a system for submarine transmission, achieved by adding a semiconductor optical amplifier (SOA) to its 10-channel transmitter and receiver PIC pair. “We are now at a point when the performance of the PIC is a good as the performance of discretes,” says Welch.
In March the company announced its next-gen PIC design - a 10x40Gbps DP-DQPSK transmitter and receiver chip pair. This is a significantly more complex design, with the transmitter integrating the equivalent of 300 optical functions; Infinera’s 10x10G transmitter PIC integrates 50.
Infinera favoured DP-DQPSK rather than the OIF-backed DP-QPSK as the latter requires significant chip support to perform the digital processing for signal recovery for each channel. Given the PIC’s 10 channels, the power consumption would be prohibitive. Instead Infinera tackles dispersion using a simpler, power-efficient optical design.
Is there a performance hit using DP-DQPSK? “There is a nominal industry figure, 1,600km reach being a good number,” says Welch. “For ultra long haul, we absolutely meet that.”
Infinera has still to launch the 400Gbps PIC whereas transponder-based system vendors have been shipping systems delivering 40Gbps lightpaths for several years. But the company says that the PIC exists, is working and all that is left is “managing it onto the manufacturing line”.
“The 40/100G area is shaping up to be the equivalent of breaking the sound barrier,” says Brad Smith, senior vice president at optical transceiver market research firm LightCounting. He questions the likely progress of optical component assemblies given they have far too many technical, cost, and size limitations. “PICs and silicon photonics have a shot at changing the game,” he says. “But the capital investment is very high with relatively low associated volumes.”
PON: an integration battleground
PON is one market where both hybrid and monolithic integration are competing with discrete-based optical transceiver designs. “Here the whole issue is cost – it’s not performance,” says Liu.
When Finisar entered the GPON transceiver market two years ago it conducted as survey as to what was available. What it found was revealing. “No-one was using the newer technologies, it was all the traditional technique based on TO cans,” said Julie Eng, vice president of optical engineering at Finisar. Mounted within the TO cans are active components such as a distributed-feedback (DFB) or Fabry-Perot laser, or a photo-detector. This is what integrated optics - whether a hybrid design basedon PLC or an indium phosphide monolithic PIC - is looking to displace.
“There is a huge infrastructure – millions of TO cans - and the challenge for hybrid and monolithic integration is that they are chasing a cost-curve that continues to come down,” says Eng.
According to NeoPhotonics’ Lipscomb, it is also hard for monolithic or hybrid integration to match the specifications of TO cans. “FTTx is similar to ROADMs, once one technology is established it is difficult for another to displace it,” he says.
But this is exactly what Canadian firms Enablence Technologies and OneChip Photonics are aiming to do.
Enablence, a hybrid integration specialist, uses a PLC-based design for PON. Onto the PLC are coupled a laser and detector for a diplexer PON design, or for a triplexer - two detectors. A common PLC optical platform is used for the different standards – Broadband PON, Ethernet PON (EPON) and Gigabit PON (GPON) - boosting unit volumes. All that is changed are the actives, for example a Fabry-Perot laser is added to the platform for 10km-reach EPON or a DFB for 20km GPON transceivers. Wavelength filtering is also performed using the PLC waveguides.
“Competing with TO cans in PON is challenging,” admits Matt Pearson, vice president of engineering at Enablence. That’s because the discrete design’s assembly is highly manual, benefiting from Far Eastern labour rates.
A hybrid approach brings several benefits, says Pearson: packaging a highly-integrated device is simpler compared to the numerous piece parts using TO cans. “It is also possible to seal at the chip level not at the module level, such that non-hermetic package can be used,” says Pearson.
There is also an additional, albeit indirect, benefit. Using hybrid integration, Enablence can reuse its intellectual property. “The same wafer process used for PON can be used for 40 and 100 Gig applications,” says Pearson. “These promise better margins as they are higher-end products.”
Enablence claims hybrid also scores when compared to monolithic integration. A hybrid design doesn’t sacrifice system performance: optimised lasers and detectors are used to meet the design specification. In contrast, performance compromises are inevitable for each of the optical functions – lasers, detectors, filtering - given that all are made in a single manufacturing process.
OneChip's EPON diplexer PIC seated on a silicon optical bench and showing the connecting fibreOneChip counters by noting the cost benefits of monolithic integration: its EPON-based transceivers are claimed to be 25 percent cheaper than competing designs. “It is not just integration [and the compact design] but there is a completely different automated packaging of the transceiver,” says Andy Weirich, OneChip’s vice president of product line management.
The company also argues that all three approaches each have their particular compromises, and that all its optical functions are high performance: the company uses a DFB laser and an optically pre-amplified photo-detector for its designs. “If you can get the best specification with no additional cost, what advantage is there of buying a cheaper laser?” says Weirich.
“Inelegant as it is, the TO can’s performance is quite good as is its cost,” says Ovum’s Liu. What is evident here is how each company is coming from a different direction, she says: “Enablence points out that a discrete design is not a platform with a future whereas the likes of Finisar are saying: do we care?”
That said, Finisar’s Eng does expect photonic integration to be increasingly used for PON: “Its time will come, we are just not at that time now.”
Photonic integration will also be used in emerging standards such as wavelength division multiplexing PON (WDM-PON), especially at the head-end where the optical line terminal (OLT) resides.
“WDM-PON is very much point-to-point even though the fibre is shared like a normal PON,” says David Smith, CTO at CIP Technologies. “When you get in the central office there is a mass of equipment just like point-to-point [access].” The opportunity is to integrate the OLT’s lasers – typically 32 or 64 - into arrays, which will also save power, says Smith.
Tunables and interconnects
Other market segments are benefiting from photonic integration besides 40 and 100Gbps transmission and PON.
JDS Uniphase’s XFP module-based tunable laser is possible by monolithically integration the laser and Mach-Zehnder modulator. Not only is the resulting tunable laser compact – it is a few millimeters long - such that it and the associated electronics fit within the module, but the power consumption is below the pluggable’s 3.5W ceiling.
JDS Uniphase has also developed a compact optical amplifier that extends long-haul optical transmission before electrical signal regeneration is needed. A PLC chip is used to replace some 50 discrete optical components including isolators, photo-detectors for signal monitoring, a variable pump splitter and tunable gain-flattening and tilt filters.
The result is an amplifier halved in size and simpler to make since the PLC removes the need to route and splice fibres linking the discretes. Moreover, JDS Uniphase can use different PLC manufacturing masks to enable specific functions for particular customers. This is the closest the optical world gets to a programmable IC.
Towards 1 terabit-per-second interfaces: a hybrid integrated prototype as part of an NIST project involving CyOptics and Kotura. Click on the photo for more details
The emerging 40 and 100 Gigabit Ethernet interface standards are another area suited for future integration. “What is driving optical integration here is size,” says Eng. “How do you fit 100Gigabit in a 3x5 inch module? That is cutting the size in half and will require a lot of R&D effort.”
In particular, optical integration will be needed to implement the 40GBASE-LR4 Gigabit Ethernet standard within a QSFP module. “You can’t fit four TO can lasers and four TO can receivers into a QSFP module,” says Rochus.
It is the 40 and 100 Gigabit Ethernet market that is the higher end market that also interests Enablence. “The drivers for PON and 100 Gigabit may be different but it’s the same PLC technology,” says Pearson. A PON diplexer may integrate one laser and one detector, for 100G it’s ten lasers and ten detectors, he says.
What next?
“Bandwidth growth is forcing us to consider architectures not considered before,” says Alcatel Lucent’s Bucci. The system vendor is accelerating its integration activities, whether it is integrating two wavelength-selective switches in a package or developing ‘electro-optic engines’ that combine advanced modulation optics and digital signal processing.
Moreover, operators themselves are more open to networking change due to the tremendous challenges they face, says Bucci: “They are being freed to do more, to take more risks.”
Welch believes one significant development that PICs will enable – perhaps a couple of years out - is adding and dropping at the packet level, at every site in the core network. “This will enable lots of reconfigurability and much finer granularity, delivering another level of networking efficiency,” he says.
Is this leading to disruption - the equivalent of digital cameras on handsets? Time will tell.
Click here for a mindmap of this article in PDF form.
Cloud computing: where telecoms and IT collide
Originally appeared in FibreSystems - May 21st 2009
IT directors worldwide are considering whether it makes financial sense to move their computing resources offsite into the 'cloud'. Roy Rubenstein assesses the potential opportunities for network operators and equipment vendors.
Cloud computing is the latest industry attempt to merge computing with networking. While previous efforts have all failed, the gathering evidence suggests that cloud computing may have got things right this time. Indeed it is set to have a marked effect on how enterprises do business, while driving the growth of network traffic and new switch architectures for the data centre.
In the mid-1990s, Oracle proposed putting computing within a network, and coined the term "network computer". The idea centred on a diskless desktop for businesses on which applications were served. The concept failed in its bid to dislodge Intel and Microsoft, but was resurrected during the dot-com boom with the advent of application service providers (ASPs).
ASPs delivered computer-based services to enterprises over the network. The ASPs faltered partly because applications were adapted from existing ones rather than being developed with the web in mind. Equally, the ASPs' business models were immature, and broadband access was in scarce supply. But the idea has since taken hold in the shape of software-as-a-service (SaaS). SaaS provides enterprises with business software on demand over the web, so a firm does not need to buy and maintain that software on its own platforms.
SaaS can be viewed as part of a bigger trend in cloud computing. Examples of cloud services include Google's applications such as e-mail and online storage, and Amazon with its Elastic Compute Cloud service, where application developers configure the computing resources they need.
Cloudy thinking
The impact of cloud starts and finishes in the IT sector. "Cloud computing is not just [for] Web 2.0 companies, it is a game-changer for the IT industry," said Dennis Quan, director of IBM's software group. "In general it's about massively scalable IT services delivered over the network."
An ecosystem of other players is required to make cloud happen. The early movers in this respect are data-centre owners and IT services companies like Amazon and IBM, and the suppliers of data-centre hardware, which include router vendors Cisco Systems and Juniper Networks, and Ethernet switch makers such as Extreme Networks and Force10 Networks.
Telecommunications carriers too are jumping on the bandwagon, which is not surprising given their experience as providers of hosting and managed services coupled with the networking expertise needed for cloud computing. International carrier AT&T, for instance, launched its Synaptic Hosting service in August 2008, a cloud-based, on-demand managed service where enterprises define their networking, computing and storage requirements, and pay for what they use. "There is a base-level platform for the [enterprise's] steady-state need, but users can tune up and tune down [resources] as required," explained Steve Caniano, vice-president, hosting and application services at AT&T.
"The top 10 operators in Europe are all adding utility-based offerings [such as storage and computing], and are moving to cloud computing by adding management and provisioning on top," said Alfredo Nulli, solutions manager for service provision at Cisco. However, it is the second- and third-tier operators in Europe that are "really going for cloud", he says, as they strive to compete with the likes of Amazon and steal a march on the big carriers.
The idea of using IT resources on a pay-as-you-go basis rather than buying platforms for in-house use is appealing to companies, especially in the current economic climate. "Enterprises are tired of over-provisioning by 150% only for equipment to sit idle and burn power," said Steve Garrison, vice-president of marketing at Force10 Networks.
Danny Dicks, an independent consultant and author of a recent Light Reading Insider report on cloud computing, agrees. But he stresses it is a huge jump from using cloud computing for application development to an enterprise moving its entire operations into the cloud. For a start-up developing and testing an application, the cost and scalability benefits of cloud are so great that it makes a huge amount of sense, he says. Once an application is running and has users, however, an enterprise is then dependent on the reliability of the connection. "No-one would worry if a Facebook application went down for an hour but it would make a big difference to an enterprise offering financial services," he commented.
The network perspective
The importance of the network and the implied demand for bandwidth as more and more applications and IT resources sit somewhere remote from the user is good news for operators and equipment makers.
If done right, there is a tremendous opportunity for telecoms operators to increase the value of their networks and create new revenue streams. At a minimum, cloud computing is expected to increase the amount of traffic on their networks.
Service providers stress the need for high-bandwidth, low-latency links to support cloud-based services. AT&T has 38 data centres worldwide, which are connected via its 40 Gbit/s MPLS global backbone network, says Gregg Sexton, AT&T's director of product development. The carrier is concentrating Synaptic Hosting applications in five "super data centres" located across three continents, linked using its OPT-E-WAN virtual private LAN service (VPLS). Using the VPLS, enterprise customers can easily change bandwidth assigned between sites and to particular virtual LANs.
BT, which describes its data centres and network as a "global cloud", also highlights the potential need for higher capacity links. "The big question we are asking ourselves is whether to go to 40 Gbit/s or wait for 100 Gbit/s," said Tim Hubbard, head of technology futures, BT Design.
Likewise, systems vendors are seeing the impact of cloud computing. Ciena first noted interest from large data-centre players seeking high-capacity links some 12 to 24 months ago. "It wasn't a step jump, more an incremental change in the way networks were being built and who was building them," said John-Paul Hemingway, chief technologist, EMEA for Ciena.
Cloud is also having an impact on access network requirements, he says. There is a need to change dynamically the bandwidth per application over a connection to an enterprise. Services such as LAN, video conferencing and data back-up need to be given different priorities at different times of the day, which requires technologies such as virtual LAN with quality-of-service and class-of-service settings.
German vendor ADVA Optical Networking has also noticed a rise in connectivity links to enterprises via demand for its FSP-150 Ethernet access product, which may be driven in part by increased demand for cloud-based services. Another area that's being driven by computing over long distances is the need to carry Infiniband natively over a DWDM lightpath. "Infiniband is used for computing nodes due to its highest connectivity and lowest latency," explained Christian Illmer, ADVA's director of business development.
Virtualization virtues
Cloud computing is also starting to influence the evolution of the data centre. One critical enabling technology for cloud computing in the data centre is virtualization, which refers to the ability to separate a software function from the underlying hardware, so that the hardware can be shared between different software usages without the user being aware. Networks, storage systems and server applications can all be "virtualized", giving end-users a personal view of their applications and resources, regardless of the network, storage or computing device they are physically using.
Virtualization enables multiple firms to share the same SaaS application while maintaining their own unique data, compute and storage resources. Virtualization has also led to a significant improvement in the utilization of servers and storage. Traditionally usage levels have been at a paltry 10 to 15%.
However, virtualization remains just one of several components needed for cloud computing. A separate management-tools layer is also needed to ensure that IT resources are efficiently provisioned, used and charged for. "This reflects the main finding of our report, that the cloud-computing world is starting to stratify into clearly defined layers," said Dicks.
Such management software can also shift applications between platforms to balance loads. An example is moving what is called a virtual machine image between servers. A virtual machine image may comprise 100 GB of storage, middleware and application software. "If [the image] takes up 5% of a server's workload, you may consolidate 10 or 20 such images onto a single machine and save power," said IBM's Quan.
Force10's Garrison notes that firms issuing request for proposals for new data centres typically don't mention cloud directly. Instead they ask questions like "Help me see how you can move an application from one rack to another, or between adjacent rows, or even between adjacent data centres 50 miles apart", he said.
Clearly, shuffling applications between servers and between data centres will drive bandwidth requirements. It also helps to explain why vendors are exploring how to consolidate and simplify the switching architecture within the data centre.
"Everything is growing exponentially, whether it is the number of servers and storage installed each year or the amount of traffic," said Andy Ingram, vice-president of product marketing and business development, data-centre business group at Juniper Networks. "The data centre is becoming a wonderful, dynamic and scary place."
This explains why vendors such as Juniper are investigating how current tiered Ethernet switching within the data centre — passing traffic between the platforms and users — can be adapted to handle the expected growth in data-centre traffic. Such growth will also strain connections between equipment: between servers, and between the servers and storage.
According to Ingram the first approach is to simplify the existing architecture. With this in mind, Juniper is looking to collapse the tiered switching from three layers to two, by linking its top-of-rack switches in a loop. Longer term, vendors are investigating the development of a singled-tiered switch in a project code-named Stratus. "We are looking to develop a scalable, flat, non-blocking, lossless data-centre fabric," said Ingram.
A flat fabric means processing a packet only once, while a non-blocking architecture removes the possibility of congestion. Such a switch fabric will scale to hundreds or even thousands of 10 Gigabit Ethernet access ports, says Ingram, who stresses that Juniper is in the first year of what will be a multi-year project to develop such an architecture.
Data centre convergence
Another development is Fibre Channel over Ethernet (FCOE) which promises to consolidate the various networks that run within a data centre. At present, servers connect to the LAN using Ethernet and to storage via Fibre Channel. This requires separate cards within the server: a LAN network interface card and a host-bus adapter for storage. FCOE promises to enable Ethernet, and one common converged network adapter card, to be used for both purposes. But this requires a new variant of Ethernet to be adopted within the data centre. Such an Ethernet development is already being referred to by a variety of names in the industry: Data Centre Ethernet, Converged Enhanced Ethernet, lossless Ethernet, and Data Centre Bridging.
Lossless Ethernet could be used to carry Fibre Channel packets since the storage protocol's key merit is that it does not lose packets. Such a development would remove one of the three main protocols in the data centre, leaving Ethernet to challenge Infiniband. But even though FCOE has heavyweight backers in the shape of Cisco and Brocade, it will probably be some years before a sole switching protocol rules the data centre.
Equipment makers believe they can benefit from the widespread adoption of cloud computing, at least in the short term. Although there will be efficiencies arising from virtualization and ever more enterprises sharing hardware, this will be eclipsed by the boost that cloud services provide to IT in general, meaning that more datacoms equipment will be sold rather than less. Longer term, however, it will probably impact hardware sales as fewer firms choose to invest in their own IT.
IBM's Quan notes that enterprises themselves are considering the adoption of cloud capabilities within their private data centres due to the efficiencies it delivers. The company thus expects to see growth of such "private" as well as "public" cloud-enabled data centres.
Dicks believes that cloud computing has a long road map. There will be plentiful opportunities for companies to deliver innovative products for cloud, from software and service support to underlying platforms, he says.
Further information
Cloud Computing: A Definition
Steve Garrison, vice-president of marketing at Force10 Networks, offers a straightforward description of cloud: accessing applications and resources and not caring where they reside.
Danny Dicks, an independent consultant, has come up with a more rigorous definition. He classifies cloud computing as "the provision and management of rapidly scalable, remote, computing resources, charged according to usage, and of additional application development and management tools, generally using the internet to connect the resources to the user."