I hadn’t been to Interop since 2003 and I’d never been to the New York show so I decided this year I would give it a shot. I was woefully disappointed. I was expecting to see great product demonstrations from all the top equipment manufacturers, but instead received inquiries for meetings from software vendors who didn’t even bother to see that I cover data centers, optical components, structured cabling and interconnects.
The exhibit hall only had eight rows. Cisco was there, but half of its booth was taken up by its channel partners and the other half had virtual demos, not actual equipment running. Brocade was there, but had a much smaller booth and pretty much legacy equipment on a tabletop display. Most telling of course were the companies that didn’t participate in the exhibition – Extreme Networks and IBM to name just two.
Some of the programming was interesting, though, and maybe made it worth the travel costs. I sat in on the second day of the Enterprise Cloud Summit so actually got to meet some of the gurus in the industry of cloud computing. I also sat in on the “Evaluating New Data Center LAN Architectures” technical session which was a panel of equipment manufacturers that responded to an RFI from Boston Scientific for a data center expansion project. Interesting to note is that while Cisco was asked to respond, it did not. The panel consisted of Alcatel-Lucent, Extreme Networks, Force10 and Hewlett Packard. It is also interesting to note that the vendors responded with different architectures – some including top-of-rack solutions and others with end-of-row.
All-in-all, I think my time would have been better spent staying home and working on my optical components research…
Monday, October 25, 2010
Thursday, October 14, 2010
SFP+ Markets
SFP+ is primarily being used in two markets currently – 10G Ethernet and 8G Fibre Channel. The Fibre Channel market is almost exclusively optical SFP+ modules while, the 10G Ethernet market is expected to see its largest growth in SFP+ direct-attach copper (DAC) products. SFP+ DAC will mainly be used in top-of-rack (ToR) switches that connect to servers within a rack. Multi-million port forecasts that are predicted over the next few years are predicated on an anticipated high-rate of adoption of this architecture in the data center. Even if this topology is used, once 10GBASE-T becomes more readily available, SFP+ DAC will be sharing its market with this less costly variant. Some OEMs believe quantities of each will be about the same, but I have my doubts. If history tells us anything, which it usually does, once the less expensive BASE-T variant takes hold, it usually pushes the more costly alternatives to much smaller volume.
But, right now it is a matter of timing. I am skeptical that the 10GBASE-T adoption in switches will be able to keep pace with the port-density need in the data center in the short-term. Both products will be needed in the long term – 10GBASE-T for inexpensive connections and 10GBASE-CR (SFP+ DAC) for lower latency, lower power consumption and more flexibility. Currently, if you want both copper and fiber ports on your 10G switch, you need to use SFP+ because there are no switches that contain both 10GBASE-T and optical.
Friday, October 8, 2010
DARPAs Chip-to-Chip Optical Interconnects (C2OI) Program
The C2OI DARPA program is funding on-going optical components projects. Its end goal is to “demonstrate optical interconnections between multiple silicon chips that will enable data communications between chips to be as seamless as data communication within a chip.”
This program grew out of work initially done by Agilent (now Avago Technologies) under the DARPA Parallel Optical Network Interconnect (PONI) project. Agilent/Avago developed a 30 Gbps transmitter (2.5 Gbps/lane) that was eventually standardized as the SNAP-12.
IBM (with help from Avago) extended the work originally done by Agilent/Avago into inter-chip connections and in 2009 achieved optical interconnection with 16 parallel lanes of 10G. By early 2010, IBM was extending this work into board-to-board applications which resulted in the new Avago MicroPOD™ product that was specifically designed for IBM’s POWER7™ supercomputer.
While it was designed for HPC server interconnects, the MicroPOD could be used for on-board or chip-to-chip interconnects as well. As mentioned in previous posts, the devices use a newly designed miniature detachable connector from US CONEC called PRIZM™ LightTurn™. The system has separate transmitter and receiver modules that are connected through a 12-fiber ribbon. Each lane supports up to 12.5 Gbps. It uses 850nm VCSEL and PIN diode arrays. The embedded modules can be used for any board-level or I/O-level application by either using two PRIZM LightTurn connectors or one PRIZM LightTurn and one MPO.
While MicroPOD is targeted at high-density HPC environments, a natural expansion of its market reach would be into switches and routers in high-density Ethernet data center environments. While this may not happen in the next few years, for me it looks like it could be a more cost-effective solution than say a 40G serial one.
This program grew out of work initially done by Agilent (now Avago Technologies) under the DARPA Parallel Optical Network Interconnect (PONI) project. Agilent/Avago developed a 30 Gbps transmitter (2.5 Gbps/lane) that was eventually standardized as the SNAP-12.
IBM (with help from Avago) extended the work originally done by Agilent/Avago into inter-chip connections and in 2009 achieved optical interconnection with 16 parallel lanes of 10G. By early 2010, IBM was extending this work into board-to-board applications which resulted in the new Avago MicroPOD™ product that was specifically designed for IBM’s POWER7™ supercomputer.
While it was designed for HPC server interconnects, the MicroPOD could be used for on-board or chip-to-chip interconnects as well. As mentioned in previous posts, the devices use a newly designed miniature detachable connector from US CONEC called PRIZM™ LightTurn™. The system has separate transmitter and receiver modules that are connected through a 12-fiber ribbon. Each lane supports up to 12.5 Gbps. It uses 850nm VCSEL and PIN diode arrays. The embedded modules can be used for any board-level or I/O-level application by either using two PRIZM LightTurn connectors or one PRIZM LightTurn and one MPO.
While MicroPOD is targeted at high-density HPC environments, a natural expansion of its market reach would be into switches and routers in high-density Ethernet data center environments. While this may not happen in the next few years, for me it looks like it could be a more cost-effective solution than say a 40G serial one.
Tuesday, October 5, 2010
DARPAs Ultraperformance Nanophotonic Intrachip Communitcations (UNIC)
UNIC is a DARPA-funded project that started in 2008 and is slated to run for about five years. Sun/Oracle, Kotura and Luxtera are working to develop this chip-to-chip "high-performance, CMOS-compatible photonic technology for high-throughput, non-blocking and power-efficient intrachip photonic communications networks."
The first application of such technology is targeted for optical interconnects for microprocessors. The goal is to replace high-performance computing clusters with computers that consist of these arrays of microprocessors interconnected by optics. Another goal of the project is to make sure the new devices are "compatible" with CMOS processes in order to also integrate the associated electronic devices. Using its now proven Silicon CMOS Photonics technology, Luxtera has developed transmitters and receivers for the project. Kotura supported the project with new low-power, high-speed modulators made of silicon photonics.
Potential new products are expected by the end of 2012. While these will be initial products, commercialization is not expected until quite some time later, perhaps not until 2016 or so. Meanwhile, Luxtera will continue to use its technology to sell 10, 40 and 100G transceivers and AOCs.
With data rates increasing beyond 10G, chip-to-chip, on-board and board-to-board optical interconnects will become progressively more significant. Even at 10G, traditional printed-circuit boards cannot support transmission beyond about 12 inches without needing re-timers. As I’ve mentioned in previous posts, instead of spending money to develop more exotic PCBs using complicated digital signal processing (DSP), it may be time to embrace optical interconnects for both board-level and chip-level.
The first application of such technology is targeted for optical interconnects for microprocessors. The goal is to replace high-performance computing clusters with computers that consist of these arrays of microprocessors interconnected by optics. Another goal of the project is to make sure the new devices are "compatible" with CMOS processes in order to also integrate the associated electronic devices. Using its now proven Silicon CMOS Photonics technology, Luxtera has developed transmitters and receivers for the project. Kotura supported the project with new low-power, high-speed modulators made of silicon photonics.
Potential new products are expected by the end of 2012. While these will be initial products, commercialization is not expected until quite some time later, perhaps not until 2016 or so. Meanwhile, Luxtera will continue to use its technology to sell 10, 40 and 100G transceivers and AOCs.
With data rates increasing beyond 10G, chip-to-chip, on-board and board-to-board optical interconnects will become progressively more significant. Even at 10G, traditional printed-circuit boards cannot support transmission beyond about 12 inches without needing re-timers. As I’ve mentioned in previous posts, instead of spending money to develop more exotic PCBs using complicated digital signal processing (DSP), it may be time to embrace optical interconnects for both board-level and chip-level.
Subscribe to:
Posts (Atom)