TechDogs-"Can IBM’s New Chip Dethrone The Industry Leader Nvidia?"

IT Infrastructure

Can IBM’s New Chip Dethrone The Industry Leader Nvidia?

By TD NewsDesk

TD NewsDesk

Updated on Mon, Aug 28, 2023

Overall Rating
The generative artificial intelligence (GenAI) race is on and as it turns out, it’s an expensive one!

Surely, you’ve come across the report showing the exorbitant operational costs by OpenAI to run ChatGPT alone. (In-depth news here!)

As per reports, the cost amounted to $700,000 per day. That’s right. Additionally, the company spent $4.6 million over two weeks to train its GPT-3 model.

Why such high prices? Well, one of the reasons is the number and price of GPUs (Graphics Processing Units) it takes to train and run such advanced technology. In the case of training its GPT-3 model, OpenAI used 9,200 GPUs. #Whaaat

It’s not just about purchasing such GPUs, it’s also about the power usage that accrues from these hungry hungry hippos. At present, the market leader in chips used for GenAI is Nvidia, with their H100 Tensor Core GPU and A100 Tensor Core GPU selling like hot cakes, making the company bet big on its product. Given the market dominance it has, their GPUs seem like a safe bet, right?

Well, it may be too soon to say, with IBM unveiling a powerful chip that shows promise of being highly energy efficient.

What’s special about the chip? Let’s explore:
 

What Did IBM Announce?

 
  • IBM announced a prototype 14nm (nanometer) analog chip that could be up to 14 times more efficient per watt than the current GPUs.

  • Presently, the chip is a research project, with a detailed research paper being published a few days ago.

  • The aim is to provide enterprises engaged in GenAI projects, like GPT-4, Midjourney and others, with reduced energy consumption and cost-efficiencies.

  • Previous simulation noted energy efficiencies ranging between 40 and 140 times as compared to currently leading GPUs.

   

How Is IBM Able To Do This?

 
  • Digital chips, which are dominant today, work with binary signals (1s and 0s).

  • IBM’s analog chips can understand the gradations between the two values by manipulating analog signals. Other varying factors include functionality, signal processing and areas of application.

  • As per a report, using phase-change memory, which consists of a material that can “switch between amorphous and crystalline phases in a way analogous to the ones and zeroes of digital processors” as well as “to states lying between these values” when hit with electrical signals.

  • Since the chip is built using analog or compute-in-memory components, the chip is capable of performing computations directly within memory.

  • Reportedly, IBM believes these 14nm chips “can encode 35 million phase-change memory devices per component, (and) can model up to 17 million parameters.”

 

How Does This Benefit AI-Centric Companies?

 
  • One of the biggest challenges today is the loss of energy and time observed while parsing large amounts of data between a processor and data. According to a report, the energy expended could range between 3 and 10,000 times the amount spent for the actual computation.

  • IBM’s analog chips could help businesses engaged in the AI industry by reducing energy usage, costs and environmental impact, while influencing the future of AI development and offering enhanced computational power.


With a few successful experiments under its belt, including a highly accurate audio transcription of people speaking, IBM’s 14nm chip could realign the future of GenAI chips. That is, if the company moves to mass produce the chips.

With Nvidia being the dominant force in the industry, will IBM be able to challenge their reign and dethrone Nvidia?

Let us know in the comments below!

First published on Mon, Aug 28, 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:

Computer ProductsIT Infrastructure Emerging Technology GenAI Generative AI Artificial Intelligence IBM Nvidia Microchips Chips Analog Digital

Join The Discussion

  • Dark
  • Light