Scientists are exploring synthetic DNA to drive the development of advanced supercomputers, addressing the growing energy demands of artificial intelligence. By combining engineered DNA sequences with quasi-2D perovskite semiconductors, experts have created memristors that mimic the brain’s synaptic plasticity for forming memories.
This approach provides exceptional data storage density, with the potential to hold 215 petabytes per gram. These devices operate at very low voltages under 0.1 volts, integrating processing and memory functions in one unit. This integration can cut energy use by up to 100 times, paving the way for scalable, efficient supercomputing systems.
Traditional computing faces thermodynamic constraints, but synthetic DNA serves as a programmable nanomaterial to overcome them. A study in the Wiley Online Library describes how doping synthetic DNA with silver ions and pairing it with perovskites forms stable conductive paths for dense storage. These memristors retain data like biological neurons, without constant power.
As AI expands, the energy required for data movement in conventional chips is becoming unsustainable. Research supported by the National Science Foundation shows that biological-inspired systems excel in parallel processing over standard architectures. DNA-based computing can handle multiple inputs with 90 percent less energy than typical non-volatile memory.
DNA’s compact nature offers storage density millions of times greater than silicon, according to National Institutes of Health studies. This could shrink data center sizes while enhancing long-term storage reliability through DNA’s chemical stability.
Although bioelectronics can be delicate, recent findings indicate that DNA-perovskite composites withstand temperatures up to 121 degrees Celsius (250 degrees Fahrenheit). This durability supports the creation of robust DNA-integrated electronics suited for the heat-intensive needs of high-performance supercomputers, potentially transforming the semiconductor field.
