Emerging Technology
Is IBM's NorthPole AI Chip The Power-Efficient Game Changer We Need?
By TechDogs Bureau
Updated on Mon, Oct 23, 2023
Share
Let's dive into the details!
IBM Corporation unveiled NorthPole, an AI chip it developed in-house, claiming to be more efficient at using power than similar CPUs in the market.
The eight years it took to create NorthPole were spent in the company's research facility in Almaden, California. The chip expands upon IBM's TrueNorth artificial intelligence processor, the result of a previous research effort. NorthPole, as claimed by IBM, is roughly 4,000 times quicker!
The newly released chip uses a 12-nanometer fabrication method to create its 22 billion transistors. IBM claims that this large number of transistors can do 2,048 eight-bit operations per clock cycle inside its 256 cores. To put it another way, the calculations are performed on numbers as little as eight bits in size.
Each of NorthPole's 256 cores has its own dedicated memory pool for storing data while it is being processed. The proximity of the chip's memory circuits and cores means that information can move quickly between them. As a result, performance is enhanced by a decrease in the processing delays normally associated with data transport.
“Architecturally, NorthPole blurs the boundary between compute and memory,” said Dharmendra Modha, IBM’s chief scientist for brain-inspired computing. “At the level of individual cores, NorthPole appears as memory-near-compute and from outside the chip, at the level of input-output, it appears as an active memory.”
What Are The Benefits Of IBM's New NorthPole AI Chip?
The memory circuits included in NorthPole's cores not only improve performance but also reduce power consumption. Transferring information from one part of a chip to another uses up resources and slows down processing. NorthPole's architecture is more efficient because it reduces the amount of time data spends traveling between memory and processing cores.
IBM determined the chip's energy usage by having it process an instance of the open-source ResNet-50 image recognition model. The model is often used to evaluate the efficacy of various AI processors. When compared to graphics cards and CPUs based on 12- and 14-nanometer nodes, IBM claims that NorthPole's power efficiency for running ResNet-50 is 25 times higher.
The NorthPole chip is manufactured at a 12-nanometer node, which is several generations behind the most modern chip manufacturing method, which operates at a 3-nanometer node. The processor, however, can be updated to make use of more recent production methods. Based on their analysis, IBM experts believe that NorthPole can improve energy efficiency by a factor of 25 compared to competing devices using "current designs."
The quantity of heat generated by a CPU is proportional to its power consumption. IBM claims that NorthPole's maximum temperature is capped by the chip's energy efficiency, meaning that it can function with minimal external cooling. Therefore, it may be used in systems with limited room for cooling equipment, such as driverless vehicles. #Innovation
With IBM's power-efficient NorthPole AI chip making waves, are we seeing the dawn of a new AI processing era?
Let us know your thoughts in the comments section below!
First published on Mon, Oct 23, 2023
Liked what you read? That’s only the tip of the tech iceberg!
Explore our vast collection of tech articles including introductory guides, product reviews, trends and more, stay up to date with the latest news, relish thought-provoking interviews and the hottest AI blogs, and tickle your funny bone with hilarious tech memes!
Plus, get access to branded insights from industry-leading global brands through informative white papers, engaging case studies, in-depth reports, enlightening videos and exciting events and webinars.
Dive into TechDogs' treasure trove today and Know Your World of technology like never before!
Disclaimer - Reference to any specific product, software or entity does not constitute an endorsement or recommendation by TechDogs nor should any data or content published be relied upon. The views expressed by TechDogs' members and guests are their own and their appearance on our site does not imply an endorsement of them or any entity they represent. Views and opinions expressed by TechDogs' Authors are those of the Authors and do not necessarily reflect the view of TechDogs or any of its officials. All information / content found on TechDogs' site may not necessarily be reviewed by individuals with the expertise to validate its completeness, accuracy and reliability.
Tags:
Related News on Emerging Technology
Are Self-Driving Cars Driving Their Own Problems?
Fri, Apr 14, 2023
By TD NewsDesk
Will Virgin Galactic Reach New Heights Or Crash?
Fri, Jun 2, 2023
By Business Wire
Oceaneering Reports Fourth Quarter 2022 Results
Fri, Feb 24, 2023
By Business Wire
Exro Announces C$30 Million Bought Deal Financing
Tue, May 16, 2023
By PR Newswire
Is LinkedIn's 1 Billion Club Your AI Career Oasis?
Fri, Nov 3, 2023
By TD NewsDesk
Join The Discussion