CS-2

Cerebras Systems Announces 130x Performance Improvement on Key Nuclear Energy Simulation over Nvidia A100 GPUs

Retrieved on: 
Monday, November 13, 2023
Software, Research, Contracts, Hardware, Nuclear, Energy, Technology, Defense, Artificial Intelligence, Semiconductor, Science, Government Technology, Other Science, GPU, Argonne National Laboratory, CSL, KAUST, SRAM, ANL, Nuclear energy, Graphics processing unit, CPU, A100, HPC, TotalEnergies, AI, National Energy Technology Laboratory, System, CS-2, Architecture, Cryptocurrency, Cerebras

Cerebras Systems , the pioneer in accelerating generative AI, today announced the achievement of a 130x speedup over Nvidia A100 GPUs on a key nuclear energy HPC simulation kernel, developed by researchers at Argonne National Laboratory.

Key Points: 
  • Cerebras Systems , the pioneer in accelerating generative AI, today announced the achievement of a 130x speedup over Nvidia A100 GPUs on a key nuclear energy HPC simulation kernel, developed by researchers at Argonne National Laboratory.
  • This result demonstrates the performance and versatility of the Cerebras Wafer-Scale Engine (WSE-2) and ensures that the U.S. continues to be the global leader in supercomputing for energy and defense applications.
  • This kernel represents the most computationally intensive portion of the full simulation, accounting for up to 85% of the total runtime for many nuclear energy applications.
  • “These published results highlight not only the incredible performance of the CS-2, but also its architectural efficiency,” said Andrew Feldman, CEO and co-founder of Cerebras Systems.

M42 Announces New Clinical LLM to Transform the Future of AI in Healthcare

Retrieved on: 
Thursday, October 12, 2023
Technology, Professional Services, Other Health, Health, Health Technology, Research, Data Analytics, Software, Artificial Intelligence, Science, Hardware, Partnership, Abu Dhabi Department of Education and Knowledge, Zawar Farid, USMLE, AI, Ethics, Data, Safety, Pharmacist, Memory, Patient, Artificial intelligence, MBZUAI, Hussain Al Nowais, CS-2, G42, ChatGPT, Literature, United States Medical Licensing Examination, Condor, LLM, Genomics, Life, Medical device, Nursing, Cerebras, M42

M42, a global tech-enabled healthcare network, has unveiled an impactful advancement in healthcare technology with the launch of Med42, a new open-access Clinical Large Language Model (LLM).

Key Points: 
  • M42, a global tech-enabled healthcare network, has unveiled an impactful advancement in healthcare technology with the launch of Med42, a new open-access Clinical Large Language Model (LLM).
  • The 70 billion parameter, generative artificial intelligence (AI) model is poised to transform the future of AI across the healthcare sector and create a direct impact on patient care outcomes.
  • The model has been developed to exponentially enhance clinical decision-making and increase access to synthesized medical knowledge for healthcare professionals, patients, and medical researchers through to regulators.
  • M42 aims to catalyze global innovation in AI for healthcare, empowering experts across the globe to refine and expand the model's applications in various medical domains.

M42 Announces New Clinical LLM to Transform the Future of AI in Healthcare

Retrieved on: 
Thursday, October 12, 2023
Partnership, Hugging Face, Abu Dhabi Department of Education and Knowledge, Research, USMLE, AI, Ethics, Data, Safety, Pharmacist, Health, Memory, Meta, Patient, Artificial intelligence, MBZUAI, ABU, Hussain Al Nowais, CS-2, G42, ChatGPT, Literature, United States Medical Licensing Examination, Condor, LLM, Genomics, Life, United Arab Emirates, Medical device, Nursing, Cerebras, M42

The 70 billion parameter, generative artificial intelligence (AI) model is poised to transform the future of AI across the healthcare sector and create a direct impact on patient care outcomes.

Key Points: 
  • The 70 billion parameter, generative artificial intelligence (AI) model is poised to transform the future of AI across the healthcare sector and create a direct impact on patient care outcomes.
  • The model has been developed to exponentially enhance clinical decision-making and increase access to synthesized medical knowledge for healthcare professionals, patients, and medical researchers through to regulators.
  • Med42 has the potential to revolutionize the pace of clinical decision-making, acting as an AI assistant for healthcare professionals.
  • Today, we helped bring another state-of-the-art LLM to the world, Med42, with huge benefits for the healthcare landscape."

Green AI Cloud and Cerebras Systems Bring Industry-Leading AI Performance and Sustainability to Europe

Retrieved on: 
Wednesday, December 14, 2022
Environment, Technology, Software, Sustainability, Alternative Energy, Energy, Internet, Hardware, Data Management, Artificial Intelligence, Cloud, Argonne National Laboratory, WSE, Tree, System, CSP, Jasper, National Center for Computational Sciences, Conference, Leibniz Supercomputing Centre, Partnership, Cloud computing, FLOPS, National Center for Supercomputing Applications, CS-2, CO2, AstraZeneca, Deep learning, GPT, EPCC, AI, Government, Absorption, Lawrence Livermore National Laboratory, European Cloud Partnership, National Energy Technology Laboratory, GSK, A100, NLP, Sandia National Laboratories, Pittsburgh Supercomputing Center, Edinburgh Parallel Computing Centre, Hydro, Carbon, HPC, Medical device, Copper, Vermilacinia cerebra, EU, Green

Cerebras Systems , the pioneer in high performance artificial intelligence (AI) compute, and Green AI Cloud , the most sustainable super compute platform in Europe, today announced the availability of Cerebras Cloud at Green AI.

Key Points: 
  • Cerebras Systems , the pioneer in high performance artificial intelligence (AI) compute, and Green AI Cloud , the most sustainable super compute platform in Europe, today announced the availability of Cerebras Cloud at Green AI.
  • A cloud provider based in the EU, such as Green AI Cloud, enables customers across the EU to benefit from Cerebras industry-leading AI compute and stay within the data privacy structures.
  • As the leader in energy efficient AI compute, it was an obvious choice to partner with and deliver AI compute to the Green AI Cloud.
  • Green AI Cloud - a European Cloud Service Provider offering AI Super Compute for the largest AI models available.

Cerebras Systems and Cirrascale Cloud Services® Introduce Cerebras AI Model Studio to Train GPT-Class Models with 8x Faster Time to Accuracy, at Half the Price of Traditional Cloud Providers

Retrieved on: 
Tuesday, November 29, 2022
Technology, Artificial Intelligence, Data Management, Cloud, Deep learning, CS-2, AI, Partnership, A100, NLP, HPC, T5, Research, GPT, Data, Workload, Cryptocurrency, Medical device, Video game console, Cerebra (British charity)

Training Large Language Models (LLMs) is challenging and expensive -- multi-billion parameter models require months to train on clusters of GPUs and a team of engineers experienced in distributed programming and hybrid data-model parallelism. It is a multi-million dollar investment that many organizations simply cannot afford.

Key Points: 
  • The Cerebras AI Model Studio offers users the ability to train GPT-class models at half the cost of traditional cloud providers and requires only a few lines of code to get going.
  • The Cerebras AI Model Studio makes this easy and dead simple just load your dataset and run a script.
  • The Cerebras AI Model Studio offers users cloud access to the Cerebras Wafer-Scale Cluster, which enables GPU-impossible work with first-of-its-kind near-perfect linear scale performance.
  • Cirrascale Cloud Services, Cirrascale and the Cirrascale logo are trademarks or registered trademarks of Cirrascale Cloud Services LLC.

Cerebras Unveils Andromeda, a 13.5 Million Core AI Supercomputer that Delivers Near-Perfect Linear Scaling for Large Language Models

Retrieved on: 
Monday, November 14, 2022
Software, Data Analytics, Internet, Hardware, Artificial Intelligence, Professional Services, Technology, Other Technology, System, Marketing, AI, United Kingdom, MSL, GPU, CS-2, Linearity, ACM, A100, Multimedia, University, Graphics processing unit, Deep learning, Argonne, Software, Nvidia, Argonne National Laboratory, Andromeda, University of Cambridge, Pain, Artificial intelligence, Memory, Project Jupyter, AMD, COVID-19, Research, Heart, Book, Genome, Medical device, Video game console, Frontier, Cerebra (British charity), Polaris

Near-perfect scaling means that as additional CS-2s are used, training time is reduced in near perfect proportion. This includes large language models with very large sequence lengths, a task that is impossible to achieve on GPUs. In fact, 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.

Key Points: 
  • It is the only AI supercomputer to ever demonstrate near-perfect linear scaling on large language model workloads relying on simple data parallelism alone.
  • Unlike any known GPU-based cluster, Andromeda delivers near-perfect scaling via simple data parallelism across GPT-class large language models, including GPT-3, GPT-J and GPT-NeoX.
  • This includes large language models with very large sequence lengths, a task that is impossible to achieve on GPUs.
  • Andromeda delivers 13.5 million AI cores and near perfect linear scaling across the largest language models, without the pain of distributed compute and parallel programing.

Boeing-Built Space Launch System Core Stage Powers First NASA Artemis Mission to Moon

Retrieved on: 
Wednesday, November 16, 2022
Space Launch System, Artemis, SPACE, Michoud Assembly Facility, Exploration Upper Stage, SLS, CS-2, Michoud, New Orleans, Safety, Kennedy Space Center Launch Complex 39, NYSE, Multimedia, Program, NASA, Culture, Kennedy Space Center, Orion, CS-4, Human, Color, Engineering, Moon, Communications satellite, Boeing

KENNEDY SPACE CENTER, Fla., Nov. 16, 2022 /PRNewswire/ -- NASA's Space Launch System rocket, powered by the Boeing [NYSE: BA] -built core stage, lifted off at 1:47 a.m.

Key Points: 
  • KENNEDY SPACE CENTER, Fla., Nov. 16, 2022 /PRNewswire/ -- NASA's Space Launch System rocket, powered by the Boeing [NYSE: BA] -built core stage, lifted off at 1:47 a.m.
  • Eight and a half minutes into flight, the core stage completed its mission and separated from the upper stage of the rocket, sending NASA's Orion spacecraft on its first journey around the Moon.
  • "Today, this country now has a super-heavy lift launch capability for the first time in 50 years," said Jim Chilton, senior vice president of Boeing's Space and Launch division.
  • Core Stage-2, or CS-2, will launch the first crew of the Artemis program and is in the final assembly area at Michoud.

Cerebras Systems and National Energy Technology Laboratory Set New Milestones for High-Performance, Energy-Efficient Field Equation Modeling Using Simple Python Interface

Retrieved on: 
Thursday, November 10, 2022
Data Management, Other Energy, Technology, Energy, Software, Artificial Intelligence, Hardware, Disruption, Work, Carbon, CS-2, United States Department of Energy, CFD, Interface, Partnership, Communication, Research, WSE, Department, ArXiv, Honda advanced technology, WFA, Computational fluid dynamics, 139th, Computer, Energy, Data science, AI, NETL, Program, HPC, Deep learning, System, Engineering, Conference, Memory, CEO, Graphics processing unit, Workload, Machine learning, API, Architecture, Central processing unit, Video game console, Computer data storage, Medical device, Joule, Cerebra (British charity)

While this performance is consistent with hand-optimized assembly codes, the WFA provides an easy-to-use, high-level Python interface that allows users to form and solve field equations effortlessly.

Key Points: 
  • While this performance is consistent with hand-optimized assembly codes, the WFA provides an easy-to-use, high-level Python interface that allows users to form and solve field equations effortlessly.
  • This work demonstrates the fastest known time-to-solution for field equations in computing history at scales up to several billion cells.
  • In the past, field equations have been memory bound, and in distributed systems, they are limited by node-to-node communication bandwidth.
  • NETL is a U.S. Department of Energy national laboratory that drives innovation and delivers technological solutions for an environmentally sustainable and prosperous energy future.

Cerebras Wafer-Scale Cluster Brings Push-Button Ease and Linear Performance Scaling to Large Language Models

Retrieved on: 
Wednesday, September 14, 2022
Technology, Semiconductor, Other Technology, Telecommunications, Hardware, Electronic Design Automation, Data Management, Artificial Intelligence, Consumer Electronics, Memory, Lawrence Livermore National Laboratory, National Center for Supercomputing Applications, Edinburgh Parallel Computing Centre, Argonne National Laboratory, Organization, GSK, EPCC, Deep learning, Government, CEO, LLM, Pain, CS-2, National Energy Technology Laboratory, NLP, AstraZeneca, Solution, Transport, Leibniz Supercomputing Centre, Natural language processing, Artificial intelligence, Project Jupyter, System, 20B, LLM Lettering, GPU, HPC, National Center for Computational Sciences, Pittsburgh Supercomputing Center, Information technology, Video game console, Typesetting, Medical device, AI, Vermilacinia cerebra

The key to the new Cerebras Wafer-Scale Cluster is the exclusive use of using data parallelism. Data parallelism is the preferred approach for all AI work. However, data parallelism requires that all the calculations, including the largest matrix multiplications of the largest layer, fit on a single device, and that all the parameters fit in the device’s memory. Only the CS-2 -- and not graphics processing units -- achieves both characteristics for LLMs.

Key Points: 
  • With a Wafer-Scale Cluster, users can distribute even the largest language models from a Jupyter notebook running on a laptop with just a few keystrokes.
  • Large language models (LLMs) are transforming entire industries across healthcare and life sciences , energy , financial services , transportation, entertainment, and more .
  • However, training large models with traditional hardware is challenging and time consuming and has only successfully been accomplished by a few organizations.
  • Instead, Cerebras Wafer-Scale Clusters deliver push-button allocation of work to compute, and linear performance scaling from a single CS-2 to up to 192 CS-2 systems.

Computer History Museum Honors Cerebras Systems with New Display for Wafer-Scale Engine

Retrieved on: 
Wednesday, August 3, 2022
Software, Hardware, Electronic Design Automation, Artificial Intelligence, Technology, Semiconductor, Arts, Museums, Entertainment, Climate change, Edinburgh Parallel Computing Centre, Argonne National Laboratory, Computer, History, CS-2, National Energy Technology Laboratory, Engineering, EPCC, Lawrence Livermore National Laboratory, National Center for Supercomputing Applications, Memory, Workload, CHM, Industry, Deep learning, National Center for Computational Sciences, Program, Leibniz Supercomputing Centre, GlaxoSmithKline, Government, Technology, Solution, Computer History Museum, Hardware acceleration, Artificial intelligence, Drug discovery, TSMC, Collection, CEO, Graphics processing unit, HPC, Pittsburgh Supercomputing Center, WSE, System, AstraZeneca, Medical device, Video game console, AI, Vermilacinia cerebra

Cerebras Systems , the pioneer in accelerating artificial intelligence (AI) compute, and the Computer History Museum (CHM), the leading institution decoding technology its computing past, digital present, and future impact on humanity today unveiled a new display featuring Cerebras Wafer-Scale Engine (WSE) .

Key Points: 
  • Cerebras Systems , the pioneer in accelerating artificial intelligence (AI) compute, and the Computer History Museum (CHM), the leading institution decoding technology its computing past, digital present, and future impact on humanity today unveiled a new display featuring Cerebras Wafer-Scale Engine (WSE) .
  • It is the honor of a lifetime to be accepted into the Computer History Museums world-renowned collection, said Andrew Feldman, CEO and co-founder of Cerebras Systems.
  • For more information on the new WSE display at the Computer History Museum, please tune into a livestream conversation on Wednesday, August 3 at 2:30 pm PT with Cerebras Systems CEO Andrew Feldman and Computer History Museum President & CEO Danl Lewin https://www.youtube.com/computerhistory/live .
  • Cerebras Systems is a team of pioneering computer architects, computer scientists, deep learning researchers, and engineers of all types.