Panel II: Pluggable Optics – How is the Ecosystem and Value Chain Changing?
Tuesday, 21 March
12:30 - 14:00
Large data centers interconnect bottlenecks are dominated by the switch I/O BW and the front panel BW as a result of pluggable transceiver modules. Recently 100G and beyond transceivers have been developed that significantly reduce power, footprint and cost for three major types of connections: intra- and inter-datacenter, transport client, and metro/access. The detailed designs of the pluggables, however, have many flavors such as: four-wave or four-fiber 28GBd NRZ, two-wave 28GBd PAM4, single-wave 56GBd PAM4, or single-wave DMT, and so on. At the same time, to overcome the front panel BW and the switch ASIC BW limitation one approach is to either move the optics onto the mid-plan or integrate the optics into the switch ASIC. There are many new MSAs in progress to be considered as well including CFP8, QSFP56, QSFP-DD, SFP56 and so on, or even chip on boards directly.
This panel of industry experts will strive to determine the potential winning technology from the wide variety of options as well as answer the following questions:
What’s the realistic price to expect for 100G pluggable to enable mass adoption?
Is the cost of a $1/Gb/s for high-speed transceivers ultimately achievable in foreseeable future and what will vendor margins look like?
How can the market serve a reach distance of 2km at one end and 40/80km at the other?
What are the lessons learned of deploying the PAM4 versus NRZ pluggables?
Will various demands of different types of data center operators be adequately accommodated with standardized vs. proprietary solutions?
When will a new 200 & 400G pluggable be commercialized and how much will it cost?
What’s the status of 100G coherent CFP2 and CFP4 pluggable modules?