Ranovus introduced its Odin Analog-Drive CPO 2.0 architecture, which leverages co-packaged Optics (CPO) to provide nx100Gbps PAM4 Optical I/O for Ethernet switch and ML/AI silicon in a single packaged assembly.
In March 2020, Ranovus announced its first generation of co-packaged optics (CPO 1.0), which enabled an attractive multi-sourced solution for 51.2T Ethernet switch applications.
CPO 2.0 improves on CPO 1.0 while maintaining optical interoperability and brings the following advantages to the ecosystem:
- 40% cost and power consumption savings through:
- Eliminating the retimer function in the Odin™ Analog-Drive optical engine
- Enabling a cost-effective single die solution with the Odin™ Analog-Drive optical engine
- Smaller footprint
- Reuse and Optimization of existing 100G PAM4 and PCIe Ser/Des chips vs new investment in an XSR Ser/Des chip for data center applications
Customer trials with Odin Analog-Drive CPO 2.0 are planned in Q4 2021.
“We launched Odin, our next generation monolithic EPIC and laser platform for multi terabit Co-Packaged Optics (CPO) applications in data centers, in March 2020. The Odin platform is a low latency, high density, protocol agnostic monolithic optical engine that delivers massive optical interconnect bandwidth with industry-leading cost and power efficiency,” said Dr. Christoph Schulien, EVP of R&D and Systems at Ranovus. “The Odin Analog-Drive configuration eliminates the retimers in the CPO 1.0 configuration, which brings about a 40% cost and power reduction to the overall system.”
Ranovus launches Silicon Photonic Engine supporting 800G
Ranovus launched its “Odin” silicon photonics device scaling from 800Gbps to 3.2Tbps in a single chip and supporting both module and Co-Packaged Optics solutions. Ranovus’ Odin platform incorporates the company’s multi-wavelength Quantum Dot Laser (QDL), 100Gbps silicon photonics-based Micro Ring Resonator modulators and photodetectors, 100Gbps Driver, 100Gbps TIA and control Integrated Circuits supported by a Tier 1 packaging ecosystem. “ML/AI…
