Data Storage Digest

Do-It-Yourself Windows File Recovery Software: A Comparison

results »

IBM Builds New Chip Based on the Human Brain

Silicon computer chips are constantly evolving, especially in the 21st century. Constant research and development is invested into the creation of smaller, faster and more efficient silicon chips, and it should come as no surprise that IBM is one of the leading pioneers in chip development today. To this extent, they've recently unveiled their new SyNAPSE technology, short for Systems of Neuromorphic Adaptive Plastic Scalable Electronics, in a brand new chip architecture that was modeled after the human brain. Sometimes referred to as the "cognitive chip," the technology, which was announced in early August 2014, is huge step forward in information technology.

The project, which has received well over $50 million in funding from multiple sources to date, is meant to pioneer a series of intelligent sensor networks that allow the chip to mimic the functionality of the human brain. Specifically, the technology is expected to reproduce human perception and cognitive abilities as seen in the human brain. IBM expects to achieve this through their new chip, the largest IBM has ever developed, which features a total of 5.4 billion separate transistors and a network of 4,096 neurosynaptic cores.

Despite the processing power contained in the cognitive chip, the device consumes a rather modest 70mW during operation, which is well less than that of the silicon chips that are on the market today. This is due to the chip's event-driven operation, which allocates processing power only as needed. In contrast, current silicon chips are constantly drawing power from your system.

Moreover, IBM's new SyNAPSE technology, as its name implies, is fully scalable, totally interconnected and able to process complex algorithms and tasks at a much quicker rate. At its current stage, SyNAPSE technology relies on a simulated software system that features multiple threads, parallel connections and a fully scalable operating environment.

IBM is dedicated to providing ongoing support for the technology through every phase, from inception to final deployment. In fact, in order to educate and train IT professionals regarding usage of the new technology, IBM is even offering curriculum based around the basic SyNAPSE framework, chip / system simulation, programming, prototype design and neuron specification.

Dharmendra S. Modha, Principal Investigator and Senior Manager with IBM Research, likened SyNAPSE technology to the FORTRAN programming language for use with synaptic silicon computer chips. He suggested that the SyNAPSE technology will pave the way for "new technological capability in terms of programming and applying emerging learning systems."

Professor Tobi Delbruck with the Institute of Neuroinformatics at UZH-ETH Zurich spoke about his partnership with IBM and the advancements they've been able to make together. He was quoted in a recent interview as saying: "We’ve been working with IBM for the last 18 months and are extremely impressed with their achievement. Applications like real time speech and vision that run continuously on battery power may finally be within reach." He ended his interview in somewhat of a pessimistic manner, saying finally: " It’s too soon to say who will win the race to implement practical realizations of brain-like computing in silicon, but IBM’s solution is a serious contender.”


No comments yet. Sign in to add the first!