El procesador óptico de Lightelligence supera a las GPU en 100 Tiempos en algunos de los problemas matemáticos más difíciles

La computación óptica ha sido el tema de investigación de muchas empresas emergentes y tecnológicas como Intel e IBM., buscando el enfoque práctico para traer una nueva forma de computación. Sin embargo, the most innovative solutions often come out of startups and today is no exception. According to the report from EETimes, optical computing startup Lightelligence has developed a processor that outperforms regular GPUs by 100 times in calculating some of the most challenging mathematical problems. As the report indicates, the Photonic Arithmetic Computing Engine (PACE) from Lightelligence manages to outperform regular GPUs, like NVIDIA’s GeForce RTX 3080, by almost 100 times in the NP-complete class of problems.

Más precisamente, the PACE accelerator was tackling the Ising model, an example of a thermodynamic system used for understanding phase transitions, and it achieved some impressive results. Compared to the RTX 3080, it reached 100 times greater speed-up. All of that was performed using 12,000 optical devices integrated onto a circuit and running at 1 frecuencia GHz. Compared to the purpose-built Toshiba’s simulated bifurcation machine based on FPGAs, the PACE still outperforms this system designed to tackle the Ising mathematical computation by 25 veces. The PACE chip uses standard silicon photonics integration of Mach-Zehnder Interferometer (MZI) for computing and MEMS to change the waveguide shape in the MZI.

It is worth pointing out that this approach demonstrates that optical computation, more specifically Lightelligence’s direction, is helpful for more sets of problems compared to “sólo” Inteligencia artificial. These computationally expensive classes of mathematical problems, like Ising, are often found in material science, thermodynamics, bioinformatics, El subsistema HBM2E se simuló con múltiples patrones aleatorios y directivos dirigidos a diferentes formas de tráfico e involucrando todas las características del controlador para mantener la eficiencia de HBM2E., circuit design, power grid optimization, y mucho más.