The performance characteristics of optoelectronic and VLSI multistage interconnection networks are compared. The bases of the comparison include speed, bandwidth, power consumption, and footprint area. The communication network used in the comparison is a synchronous packet-switched multistage interconnection network built from 2 × 2 bit-serial switching elements. CMOS technology is used in the VLSI implementation, and it is assumed that the entire network resides on a single chip. Regular free-space optical interconnects are used in the optoelectronic implementation and it is assumed that modulators and detectors are integrated with electronic circuitry. Both implementations rely on existing widely used components and design methods. The results show that for large networks optoelectronics offers higher speed and lower area than VLSI. The speed of the optoelectronic network remains constant as the network grows, while the VLSI network speed diminishes as O(1/N due to wire delay. The area of the optoelectronic network grows as O(N · log N), while the VLSI network area grows as O(N2· log N). The volume of the optoelectronic network grows as O(N3/2 · log N). Based on the assumed technology parameters, optoelectronics outperforms VLSI in bandwidth for network sizes above 256. The methodology developed in this paper can be applied to other communication networks and technologies.
- interconnection networks
- optical interconnects
- optoelectronic computing
ASJC Scopus subject areas
- Atomic and Molecular Physics, and Optics