Twenty-some years ago, during the heady days of
the dot-com bubble, the optical networking industry was scrambling to lace the globe with optical fiber components to meet an anticipated explosion in
bandwidth demand. The arguably irrational
exuberance and speculation during this time made the Optical Fiber Conference (OFC) the most dazzling industry trade show to attend for an ingenuous new editor — and dazzling in every sense of the word.
OFC was both overwhelming and theatrical. The overwhelm came from its scale and scope, while the theater was provided by the methods exhibitors employed to attract people into their booths. I recall one booth featuring aerialists performing a breathtaking ribbon dance high above our heads.
The dot-com crash tanked speculation in the value of online assets. Investment in optical networking merely returned to earth. The worldwide build-out of dark fiber continued, as did sober discussions about the technical and logistical challenges
of using photons to transmit information — only now these
discussions occurred without the distraction of circus
performances in the background.
After the optical networking industry cast its net around the globe, it spent the ensuing decades making the globe seem smaller. Video calls now bring intercontinental business teams into the same room. Music, television, and movies from every culture can be streamed on demand. And, as the global
pandemic unfolded, international conference-goers could attend events virtually.
Our shrinking planet is an illusion made possible by advanced optical data center technologies able to transmit data faster. Among them are incumbent direct-detect Pam4 modules and cutting-edge coherent optics capable of pushing data rates to 1.2 Tbit/s and beyond. The correlation between “smaller” and “faster” extends to the increasingly compact silicon photonics platforms that are enabling these accelerating data rates.
The trajectory of silicon photonics is not as compelling a
narrative as the fiber build-out two decades ago. It evokes less
of the ruthless railroad baron and more of the quotidian
expansion of the semiconductor industry. But, like
semiconductor manufacturing, what silicon photonics lacks in narrative power, it makes up for in its potential to revolutionize industries including communications, automotive, medicine,
and consumer electronics.
Unlike semiconductors, silicon photonics lacks an equivalent
for Moore’s Law. The sector also lacks broadly accepted standards, or roadmaps, or even a firm commitment to silicon. Material platforms, such as indium phosphide, silicon nitride, thin-film lithium niobate, and others, offer competitive alternatives, comparative limitations, and complementary capabilities.
All these factors present an overwhelming array of questions,
challenges, and opportunities with respect to the future of integrated photonics. With this overwhelm, some of the sector’s circus dazzle has also returned. There are rumors that OFC’s Rump Session on silicon photonics will include live lions.
The challenges and opportunities of silicon photonics will also be the subject of an off-site panel discussion during OFC that SOITEC is hosting at the Hard Rock Cafe in San Diego. Scheduled for March 8 and moderated by this editor, the panel will bring together four industry leaders to discuss the challenges
and outlook of heterogenous silicon photonics platforms as
they apply to next-generation datacom and other applications. Stay tuned to my LinkedIn page or SOITEC’s for further details. We’ll see you at the show.
