CHANNELS

Aerospace

Automotive

Automated & Autonomous Vehicles

Battery & Electrification Technology

Defense

Drone TV

Electronics

Manufacturing & Prototyping

Materials

Medical

Motion Control

RF & Microwave Technology

Robotics & Automation

Sensors & IoT

Test & Measurement

Aerospace
Automotive
Automated & Autonomous Vehicles
Battery & Electrification Technology
Defense
Drone TV
Electronics
Manufacturing & Prototyping
Materials
Medical
Motion Control
RF & Microwave Technology
Robotics & Automation
Sensors & IoT
Test & Measurement
Aerospace & Defense
Search Results

Brain-Like Chip Helps Computers Solve Complex Cognitive Tasks

Lawrence Livermore National Laboratory will receive a first-of-a-kind brain-inspired supercomputing platform for deep learning developed by IBM Research. Based on a breakthrough neurosynaptic computer chip called IBM TrueNorth, the scalable platform will process the equivalent of 16 million neurons and 4 billion synapses and consume the energy equivalent of a hearing aid battery - only 2.5 watts of power. The brain-like, neural network design of the IBM Neuromorphic System is able to infer complex cognitive tasks such as pattern recognition and integrated sensory processing far more efficiently than conventional chips. The new system will be used to explore new computing capabilities important to the National Nuclear Security Administration's missions in cybersecurity, stewardship of the nation's nuclear weapons stockpile, and nonproliferation.

Topics:
Computers Defense Electronic Components Electronics & Computers Semiconductors & ICs

More From SAE Media Group

    Transcript

    00:00:04 lawrence livermore is starting to work with IBM in a partnership in collaboration to explore how the IBM TrueNorth neuromorphic chip is going to be able to be applied to problems in the national interest specifically we are looking to see how we can tackle simulation modeling problems and pattern recognition classification inference problems across

    00:00:26 the national security space but in particular in the fields that the ASC or advanced simulation and computing program is tasked to explore and develop the key innovations that make to North interesting are that it's very low-power it operates at speeds close to real times it is a very large spatially distributed architecture and it uses asynchronous communication techniques to

    00:00:54 allow the system to try and solve the problem in some ways at its own pace very much the way the human brain does as we push HPC architectures towards exascale one of the main challenges we are facing is how much energy is consumed by the system when it's trying to run the computations we're interested in right now we're applying many tools to help alleviate that that's

    00:01:18 high-performance well-designed traditional CPUs bringing in GPUs tightly coupling them with the CPUs and non-volatile memory the TrueNorth architecture is a fourth approach that we're going to integrate into HPC systems that will allow us to solve certain class of applications with much less energy required and thus able to bring the entire system energy down as

    00:01:40 we mature this technology and research into new ways to use this this is going to allow us add a new class of accelerator to these large HPC systems fundamentally expanding their capability without dramatically increasing their power budget TrueNorth is a new approach to computing architecture that will allow us to innovate in how we solve algorithms and also how we develop

    00:02:03 high-performance computing architectures we're looking forward towards exascale where we're going to see heterogeneous systems that include hyperforin CPUs GPUs non-volatile memory hopefully neuromorphic chips such as Sri North this is a really exciting development and mirrors what we've seen with the rise of graphics processing units or GPUs over the past decade as

    00:02:24 they've gone from gaming systems into scientific computing with this collaboration with IBM and working with the TrueNorth architecture we're looking to change how we fundamentally solve large-scale scientific computing and simulation problems both in the interests of basic science and advancing the state of the art as well as solving critical problems in the mission space

    00:02:45 of national security