TeqGo.com
No Result
View All Result
No Result
View All Result
TeqGo.com
No Result
View All Result
Home Computer

Hungry for AI? New supercomputer contains 16 dinner-plate-size chips

Staff by Staff
November 14, 2022
in Computer
0
Hungry for AI? New supercomputer contains 16 dinner-plate-size chips
467
SHARES
1.5k
VIEWS
Share on FacebookShare on Twitter


Enlarge / The Cerebras Andromeda, a 13.5 million core AI supercomputer.

On Monday, Cerebras Systems unveiled its 13.5 million core Andromeda AI supercomputer for deep learning, reports Reuters. According Cerebras, Andromeda delivers over one 1 exaflop (1 quintillion operations per second) of AI computational power at 16-bit half precision.

The Andromeda is itself a cluster of 16 Cerebras C-2 computers linked together. Each CS-2 contains one Wafer Scale Engine chip (often called “WSE-2”), which is currently the largest silicon chip ever made, at about 8.5-inches square and packed with 2.6 trillion transistors organized into 850,000 cores.

Cerebras built Andromeda at a data center in Santa Clara, California, for $35 million. It’s tuned for applications like large language models and has already been in use for academic and commercial work. “Andromeda delivers near-perfect scaling via simple data parallelism across GPT-class large language models, including GPT-3, GPT-J and GPT-NeoX,” writes Cerebras in a press release.

The Cerebras WSL2 chip is roughly 8.5-inches square and packs 2.6 trillion transistors.
Enlarge / The Cerebras WSL2 chip is roughly 8.5-inches square and packs 2.6 trillion transistors.

Cerebras

Advertisement

The phrase “Near-perfect scaling” means that as Cerebras adds more CS-2 computer units to Andromeda, training time on neural networks is reduced in “near perfect proportion,” according to Cerebras. Typically, to scale up a deep-learning model by adding more compute power using GPU-based systems, one might see diminishing returns as hardware costs rise. Further, Cerebras claims that its supercomputer can perform tasks that GPU-based systems cannot:

GPU impossible work was demonstrated by one of Andromeda’s first users, who achieved near perfect scaling on GPT-J at 2.5 billion and 25 billion parameters with long sequence lengths—MSL of 10,240. The users attempted to do the same work on Polaris, a 2,000 Nvidia A100 cluster, and the GPUs were unable to do the work because of GPU memory and memory bandwidth limitations.”

Whether those claims hold up to external scrutiny is yet to be seen, but in an era where companies often train deep-learning models on increasingly large clusters of Nvidia GPUs, Cerebras appears to be offering an alternative approach.

How does Andromeda stack up against other supercomputers? Currently, the world’s fastest, Frontier, resides at Oak Ridge National Labs and can perform at 1.103 exaflops at 64-bit double precision. That computer cost $600 million to build.

Access to Andromeda is available now for use by multiple users remotely. It’s already being utilized by commercial writing assistant JasperAI and Argonne National Laboratory, and the University of Cambridge for research.



Source link

Previous Post

Nike Will Let People Design and Sell Sneakers for the Metaverse

Next Post

Apple Sued for Allegedly Collecting User Data, Violating Its Own Privacy Policies

Next Post
Apple Sued for Allegedly Collecting User Data, Violating Its Own Privacy Policies

Apple Sued for Allegedly Collecting User Data, Violating Its Own Privacy Policies

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Recommended.

How hard is it to get time off at Starbucks?

August 24, 2022
Remote Security

The Benefits of Remote Monitoring for Your Security Camera System

May 8, 2023

Trending.

What happened in the year 0 AD?

August 23, 2022

What is a 100000 year period called?

August 23, 2022

What does partner of the quarter mean Starbucks?

August 24, 2022

Who are the new Fox Weather anchors?

August 24, 2022

Who left Fox 2 Detroit?

August 24, 2022
  • About us
  • Contact Us
  • Home
  • Privacy Policy and Disclaimer

© 2021-2023 Teqgo.com

No Result
View All Result
  • About us
  • Contact Us
  • Home
  • Privacy Policy and Disclaimer

© 2021-2023 Teqgo.com