IISc researchers recently developed a prototype of an analog chipset called “ARYABHAT-1”.
Highlights
- The team has created a design framework to develop next-generation analog computing chipsets.
- These chipsets may operate faster. It will use less power as compared to digital processors used in different electronic gadgets.
- It has been designed by Pratik Kumar, who is a Ph.D. student at IISc.
About ARYABHAT-1
- ARYABHAT-1stands for “Analog Reconfigurable Technology and Bias-scalable Hardware for AI Tasks”.
- Such chipsets can be beneficial for applications based on Artificial Intelligence(AI) such as object or speech recognising apps including Alexa. It can also be beneficial for applications requiring very efficient parallel computations.
- Digital chips are used in many electronic devices, especially those require computers because design process is scalable and straightforward.
Benefits of Analog Computing
Analog computing is potentially active to perform as compared to digital computing in applications which do not call for accurate computations. Analog computing is more energy-efficient.
Challenges of making analog chips
Multiple technological challenges are associated with building analog chips:
- Testing and co-designing the Analog processor is challenging, as compared to that of digital chips. Large-scale digital processors can be developed quickly by compiling high-level code.
- Analog chips are difficult to scale as well. They are specially tailored while shifting to new applications or next generation of technology.
- It would not be easy to trade-off power and area for the purpose of speed and accuracy. Though, accuracy can be improved by incorporating additional elements on same chip, like logic units.
Applications of ARYABHAT
According to researchers, ARYABHAT is capable to be configured with several machine learning architectures. For instance, with the digital CPUs. It has the potential to function reliably on different temperature range.