INDUSTRY PHOTONIC INTEGRATION
IT TAKES JUST A YEAR and a half for the bandwidth demand for telecommunications networks to double. The primary drivers for this are the soaring number of networked devices – half of Americans now have smart phones – and an increase in data downloaded from the likes of Netflix, Amazon and iTunes.
Rocketing levels of data transfer don’t just strain the telecom networks – they also create an unprecedented challenge for their datacom siblings. Lying at the heart of these datacom networks are datacenters, which are used for a diverse range of applications and can often feature tens of thousands of servers. These servers require flexible, agile interconnect networks, which must be capable of allocating jobs to any server, at any time, to respond to user demand.
The telecom and datacom networks, which work together to control the flow of internet traffic, place different demands on the evolution of the underlying interconnect technology. Long-haul telecom technology is optimised for spectral efficiency and high bandwidth, due to the cost of deploying fibers in the field. In comparison, short- reach datacom technology offers a cost-effective, low-power approach for connecting extremely large numbers of data ports in a single location. These differences account for the need for: high- performance components in telecom networks; and for lower-cost components, which can be made in far higher volumes, in datacom networks.
Scaling issues
Products for serving both of these networks must deliver increases in performance as their dimensions are reduced. This performance- scaling requirement suggests that the only viable route ahead is a chip-based approach, rather than conventional discrete component assembly. This should cater for increases in the bandwidth requirements per transceiver, along with its complexity and component count.
It is tough to scale existing photonic technologies for datacom networks, because next-generation products have to be cheaper, produced in higher volumes, run off less power, and be housed in smaller packages. On the long-haul side of the industry, the focus has been on developing high- performance transceivers based on complex, InP-based photonic integrated circuits (PICs). These chips have satisfied performance requirements, but if they are to meet the cost and volume needs of datacentre interconnects, a significant ramp in manufacturing capability must occur. Today, the incumbent technology in that sector is a combination of VCSELs and
July 2013
www.compoundsemiconductor.net 37
multi-mode fiber. This product serves low-cost, high-volume markets by delivering high aggregate bandwidths, but it lacks sufficient reach for large datacentres.
In contrast, silicon photonics – a relatively new area of the photonics industry – promises to provide a platform that will scale to fulfill the demands of next-generation datacom interconnects. Its attractive features include: the ability to digitally enhance photonics through co-design of optical devices in silicon with electronics; the leveraging of high-precision, shared foundries using wafers with diameters of up to 12-inches; and silicon’s compatibility with the developments in advanced packaging. The latter attribute means that there is potential to move photonics out of a ‘gold box’ package and into one with intimate interconnection to the electronic chip.
Despite all its potential, commercial realization of silicon photonics is taking longer than many had hoped for. Up until now, the laser has been segregated from the rest of the chip, leading to compromised system architecture: The laser is then treated like a power supply and split into many parallel channels, leading to a hike in cabling costs, rather than utilising the spectrally allocated bandwidth provided by wavelength division multiplexing (WDM) systems.
One upshot of separating the laser from the rest of the chip is that it delays penetration of silicon photonics into datacom markets. Performance is also compromised, preventing this technology from impacting most telecom markets. However, early work in heterogeneous integration by groups at Intel, Ghent University and the University of
Figure 1. Relative bandwidth growth for different sectors normalized to the year 2010. Source: IEEE 802.3 BWA Ad Hoc Report, 19th July 2012
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140 |
Page 141 |
Page 142 |
Page 143 |
Page 144 |
Page 145 |
Page 146 |
Page 147 |
Page 148 |
Page 149 |
Page 150 |
Page 151 |
Page 152 |
Page 153 |
Page 154 |
Page 155 |
Page 156 |
Page 157 |
Page 158 |
Page 159 |
Page 160 |
Page 161 |
Page 162 |
Page 163 |
Page 164 |
Page 165 |
Page 166 |
Page 167 |
Page 168 |
Page 169 |
Page 170 |
Page 171 |
Page 172 |
Page 173 |
Page 174 |
Page 175 |
Page 176 |
Page 177 |
Page 178 |
Page 179