Photonic Memory & Storage: A Paradigm Shift for Next-Generation Computing

The rapid growth of data in recent years due to modern technologies like Artificial Intelligence, IoT, Cloud computing, big data analytics, etc., has put a lot of pressure on modern storage and processing systems. Current e-memory systems like DRAM and NAND flash storage lag behind in the market demands of faster operating speed, better efficiency, and greater scalability. These limitations have originated from the first principles associated with electronic charge-based storage, such as energy dissipation, latency and bandwidth limitations. When extending computing architectures towards the exascale level and neuromorphic computing, new memory paradigms are needed to close the gap between data processing and data storage. Such technologies include photonic memory and storage; this is because photonic memory and storage is a relatively new field of computation that uses light-based methods of storing data. Photonic systems contrast with conventional electronic counterparts in that they involve much less power consumption during standby conditions, perform data accesses with high speed, and cannot be affected by electromagnetic interferences. Optical memory devices are, therefore, based on this concept of optically encodable data, light pulses or Phase Change Materials (PCMs). Some of the materials, like chalcogenide-based phase change compounds, have the ability to undergo phase change both in optical and structural characteristics, providing a solid basis for optoelectronic applications like non-volatile storage with high switching rates. These can also be coupled with silicon photonics, which ensures that they are fully compatible with the current semiconductor processing technology. Furthermore, the architecture of photonic memory has multi-level storage, high endurance and stability in the long run than the charge storage. As a novel leading-edge technology for data centres, HPC, and AI applications, photonic memory stands capable of changing the trends of Information technology systems. In this paper, the background highlights the fundamentals, materials, and technologies based on one and two-port device architectures and real-world applications of PM. It presents a review of the state-of-the-art of its implications for future computing and data storage