Cerebras launches new AI Supercomputing Processor, WSE-2: 2.6 Trillion Transistors, 100% Yield

Within the last few years, we have seen a huge amount of processors enter the market and their sole purpose is to accelerate artificial intelligence and machine learning workloads. These processors are often focused on a few key areas, due to the different types of machine learning algorithms possible, but the size limits them all.

Two years ago, a computer systems company, Cerebras, unveiled a revolution in silicon design: a processor as big as a human head, built on 16nm, using as much area on a 12-inch wafer as a rectangular design would allow, focused on both HPC workloads and AI as well. On 20th April, the company launched its second-generation product, built on TSMC 7nm, with more than double the cores and pretty much more than double of everything.

Second Generation Wafer Scale Engine

Cerebras builds the new processor on the first by moving to TSMC’s N7 process. This allows SRAMs to scale down to some extent, and the logic to scale down, and now the new chip has 850,000 AI cores on board. Basically, almost everything about the new chip is over 2 times.

image 30 Cerebras launches new AI Supercomputing Processor, WSE-2: 2.6 Trillion Transistors, 100% Yield
Source: Anand Tech

The original processor is known as the Wafer Scale Engine (WSE-1), and the new one is named WSE-2. The WSE-2 features hundreds of thousands of AI cores across a massive 46225 mm2 of silicon.  In that space, 2.6 trillion transistors for 850,000 cores have been enabled by Cerebras – by comparison, the second biggest AI CPU on the market is ~826 mm2, with 0.054 trillion transistors, according to Anand Tech. Cerebras also cites 1000x more onboard memory, with 40 GB of SRAM, compared to 40 MB on the Ampere A100.

2D Mesh with FMAC datapaths is connected with the cores. With WSE, Cerebras’ goal is to provide a single platform, “designed through innovative patents, that allows for bigger processors useful in AI calculations but has also been extended into a wider array of HPC workloads.”

Cerebras launches new AI Supercomputing Processor, WSE-2: 2.6 Trillion Transistors, 100% Yield

Building on First Gen WSE

The custom graph compiler is a key to the design. The compiler takes pyTorch or TensorFlow and maps each layer to a physical part of the chip. This allows for asynchronous computation as the data flows through. Having such a large processor means the data can continually be moved onto the next stage of the calculation as it never has to go off-die and wait in memory, wasting power. Sparsity has been kept in mind while designing the compiler and processor, allowing high utilization regardless of batch size, or can enable parameter search algorithms to run simultaneously.

Cerebras%20WSE2%20Launch%20Embargoed%20until%204%2020%2021%5B2%5D%5B2%5D%5B3%5D%5B3%5D%5B4%5D%5B1%5D page Cerebras launches new AI Supercomputing Processor, WSE-2: 2.6 Trillion Transistors, 100% Yield

WSE-1 is sold as a complete system called CS-1, and several dozen customers with deployed systems up and running are present, including a number of pharmaceutical companies, research laboratories, military, biotechnology research, and the oil and gas industries. “Lawrence Livermore has a CS-1 paired to its 23 PFLOP ‘Lassen’ Supercomputer. Pittsburgh Supercomputer Center purchased two systems with a $5m grant, and these systems are attached to their Neocortex supercomputer, allowing for simultaneous AI and enhanced compute.”

Cerebras%20WSE2%20Launch%20Embargoed%20until%204%2020%2021%5B2%5D%5B2%5D%5B3%5D%5B3%5D%5B4%5D%5B1%5D page Cerebras launches new AI Supercomputing Processor, WSE-2: 2.6 Trillion Transistors, 100% Yield

The uniqueness of Cerebras’ design is being able to go beyond the reticle limit, the physical manufacturing limits normally presented in manufacturing. As connecting two areas with a cross-reticle connection is difficult, processors are designed with this reticle limit as the maximum size of a chip.

Cerebras remains the only one offering a processor on this scale – “the same patents that Cerebras developed and were awarded to build these large chips are still in play here, and the second-gen WSE will be built into CS-2 systems with a similar design to CS-1 in terms of connectivity and visuals.”

Cerebras%20WSE2%20Launch%20Embargoed%20until%204%2020%2021%5B2%5D%5B2%5D%5B3%5D%5B3%5D%5B4%5D%5B1%5D page Cerebras launches new AI Supercomputing Processor, WSE-2: 2.6 Trillion Transistors, 100% Yield

Cerebras states that having the solution to such a large single-chip means that the barrier to distributed training methods across 100s of AI chips is now so much further away, that this excess complication is not needed in most scenarios – to that, we’re seeing CS-1 deployments of single systems attached to supercomputers. However, the company is keen to point out that “two CS-2 systems will deliver 1.7 million AI cores in a standard 42U rack, or three systems for 2.55 million in a larger 46U rack (assuming there’s sufficient power for all at once!), replacing a dozen racks of alternative computer hardware.”

At Hot Chips 2020, Sean Lie, Chief Hardware Architect of Cerebras, stated that one of the key benefits to customers of the company was the ability to enable workload simplification that previously required racks of GPU/TPU but instead can run on a single WSE in a computationally relevant fashion.

Cerebras%20WSE2%20Launch%20Embargoed%20until%204%2020%2021%5B2%5D%5B2%5D%5B3%5D%5B3%5D%5B4%5D%5B1%5D page Cerebras launches new AI Supercomputing Processor, WSE-2: 2.6 Trillion Transistors, 100% Yield

As a company, Cerebras has ~300 staff across different countries including Canada, the USA, and Japan. They have dozens of customers already with CS-1 deployed and a number more already trialing CS-2 remotely as they bring up the commercial systems. 

SOURCE

LEAVE A REPLY

Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.

More like this

TSMC Opens 2nm Wafer Orders from April 1: Apple Likely First in Line

TSMC Opens 2nm Wafer Orders from April 1: Apple...

Taiwan Semiconductor Manufacturing Company (TSMC) is set to begin accepting orders for its highly anticipated 2nm wafers...
TSMC 2nm Trial Production

TSMC 2nm Trial Production Yields Surpass 60% Milestone —...

TSMC 2nm Trial Production : Taiwan Semiconductor Manufacturing Company (TSMC) has reportedly made impressive strides in...
Tensor G5 for Pixel 10: Google’s First Fully Custom ISP & TSMC’s 3nm Process

Tensor G5 for Pixel 10: Google’s First Fully Custom...

Google is making a significant shift with the Tensor G5 chip for the upcoming Pixel 10, marking...
Intel’s Comeback: How the Trump Administration, TSMC & Broadcom Are Reshaping the Foundry Race

Intel’s Comeback: How the Trump Administration, TSMC & Broadcom...

Intel's foundry business, once a symbol of semiconductor dominance, is on the brink of a massive revival—thanks...
TSMC Races Toward 1nm Breakthrough: Plans for Giga Fabs in Taiwan Unveiled

TSMC Races Toward 1nm Breakthrough: Plans for Giga Fabs...

The semiconductor industry is on the brink of another revolution, and Taiwan Semiconductor Manufacturing Company (TSMC) is...

LATEST NEWS

The Last of Us Part 2: Ellie’s Age and Evolution – From Teen Survivor to Hardened Warrior

In the haunting, post-apocalyptic world of The Last of Us, few characters have captured our hearts and imaginations quite like Ellie. From the moment...

Assassin’s Creed Shadows: Mastering the Chef Hong Quest – A Culinary Adventure in Feudal Japan

In the heart of feudal Japan, where shadows conceal both danger and opportunity, Assassin's Creed Shadows offers players a unique blend of stealth, combat,...

Snapdragon 8s Gen 4 Phones: iQOO, Redmi, Poco & More Launching Soon

Qualcomm has officially introduced the Snapdragon 8s Gen 4 chipset, following the Snapdragon 8s Gen 3. Though there was a rumor that the "Elite"...

Samsung’s AI Revolution: Galaxy A Series Now Packs “Awesome Intelligence” Features

Samsung’s AI Revolution: Remember when cutting-edge AI features were exclusively for those willing to spend $1000+ on a flagship smartphone? Those days are officially...

Featured