OE-CAM: A Hybrid Opto-Electronic Content Addressable Memory

Abstract

A content addressable memory (CAM) is a type of memory that implements a parallel search engine at its core. A CAM takes as an input a value and outputs the address where this value is stored in case of a match. CAMs are used in a wide range of applications including networking, cashing, neuromorphic associative memories, multimedia, and data analytics. Here, we introduce a novel opto-electronic CAM (OE-CAM) utilizing the integrated silicon photonic platform. In our approach, we explore the performance of an experimental OE-CAM and verify the efficiency of the device at 25Gbit/s while maintaining the bit integrity under noise conditions. We show that OE-CAM enables a) two orders of magnitude faster search functionality resulting in b) a five orders of magnitude lower power-delay-product compared to CAMs implementations based on other emerging technologies. This remarkable performance potential is achieved by utilizing i) a high parallelism of wavelength-divisionmultiplexing in the optical domain, combined with ii) 10's of GHz-fast opto-electronic components, packaged in iii) integrated photonics for 10-100's ps-short communication delays. We further verify the upper optical input power limit of this OE-CAM to be given by parasitic nonlinearities inside the silicon waveguides, and the minimal detectable optical power at the back-end photoreceiver's responsivity given channel noise. Such energy-efficient and short-delay OE-CAMs could become a key component of functional photonic-augmented ASICS, co-processors, or smart sensors.

Strain-engineered high-responsivity MoTe2 photodetector for silicon photonic integrated circuits

“…Photodetectors are universal building blocks for i) data links and interconnects including hybridtechnology options [48] including dual-function optoelectronic devices [49], ii) photonic integrated network-on-chip technology [50], iii) photonic digital-to-analog converters [51], and photonic analog processors such as iv) photonic residue arithmetic processors relying on crossbar routers [52,53], v) analog partial differential equation solvers [54], vi) photonic integrated neural networks [55][56][57][58], and vii) photonic reprogrammable circuits such as content-addressable memories [59].…”

Computing of neuromorphic materials: an emerging approach for bioengineering solutions

“…Different signal kinds or computing tasks may respond better to various encoding strategies. [195][196][197] Neuromorphic computing also requires a careful consideration of the degree of the complexity of the encoding scheme, as well as the hardware resources available for implementing the decoding operations. Ultimately, the choice of encoding and decoding techniques should be tailored to the specific computing task and the available hardware resources.…”

Silicon-Based Group-IV O-E-O Devices for Gain, Logic, and Wavelength Conversion

“…25 To supplement the EP logic circuits, the memory portion of the EPDC can be enhanced by using photonic components to create an electronic-photonic content addressable memory. 26 In summary, the EPDC has the potential to advance DOC in a major way.…”

“…In cases where the logic circuit is only EOL, the OEO “repeater” devices are readily inserted into the circuit to boost signals and to reduce the bit error rate . To supplement the EP logic circuits, the memory portion of the EPDC can be enhanced by using photonic components to create an electronic-photonic content addressable memory . In summary, the EPDC has the potential to advance DOC in a major way.…”