Low-power 'microwave brain' on a chip computes on both ultrafast data and wireless signals

Amazing stuff!

"... researchers have developed a low-power microchip they call a "microwave brain," the first processor to compute on both ultrafast data signals and wireless communication signals by harnessing the physics of microwaves. ..."

the processor is the first true microwave neural network and is fully integrated on a silicon microchip. It performs real-time frequency domain computation for tasks like radio signal decoding, radar target tracking and digital data processing, all while consuming less than 200 milliwatts of power. ..."

From the abstract:

"The development of high-bandwidth applications, including multi-gigabit communication and radar imaging, demands faster processing. However, in the microwave regime, where frequencies exceed clock rates, sampling and computation become challenging.

Here we report an integrated microwave neural network for broadband computation and communication. Our microwave neural network operates across tens of gigahertz but is reprogrammed with slow megabits per second control bitstreams.

By exploiting strong nonlinearity in coupled microwave oscillations, it expresses its computation in a narrower spectrum, enabling easy read-out. The system searches bit sequences in multi-gigabits per second data and emulates digital functions without custom circuits. It accelerates radio-frequency machine learning by classifying encoding schemes and detecting frequency shifts to track flight trajectories from radar.

The microwave neural network is fabricated with standard complementary metal–oxide–semiconductor technology. It occupies a sub-wavelength footprint of 0.088 mm2 on chip and has a sub-200-mW power consumption, supporting integration in a general-purpose analogue processor."

Low-power 'microwave brain' on a chip computes on both ultrafast data and wireless signals

An integrated microwave neural network for broadband computation and communication (no public access)

An integrated microwave neural network for broadband computation and communication (preprint, open access)

Nvidia Unveils Blackwell, Its Next GPU

Good news!

"Today [3/17/2023] at Nvidia’s developer conference, GTC 2024, the company revealed its next GPU, the B200. The B200 is capable of delivering four times the training performance, up to 30 times the inference performance, and up to 25 times better energy efficiency, compared to its predecessor, the Hopper H100 GPU. Based on the new Blackwell architecture, the GPU can be combined with the company’s Grace CPUs to form a new generation of DGX SuperPOD computers capable of up to 11.5 billion billion floating point operations (exaflops) of AI computing using a new, low-precision number format. ...

The B200 is composed of about 1600 square millimeters of processor on two silicon dies that are linked in the same package by a 10 terabyte per second connection, so they perform as if they were a single 208-billion-transistor chip. Those slices of silicon are made using TSMC’s N4P chip technology, which provides a 6 percent performance boost over the N4 technology used to make Hopper architecture GPUs, like the H100.

Like Hopper chips, the B200 is surrounded by high-bandwidth memory, increasingly important to reducing the latency and energy consumption of large AI models. B200’s memory is the latest variety, HBM3e, and it totals 192 GB (up from 141 GB for the second generation Hopper chip, H200). Additionally, the memory bandwidth is boosted to 8 terabytes per second from the H200’s 4.8 TB/s. ..."

Nvidia Unveils Blackwell, Its Next GPU - IEEE Spectrum

NVIDIA Blackwell Platform Arrives to Power a New Era of Computing

Old and new processor side by side

This Is How China's Huawei Shocked America With a Smartphone

Very recommendable!

Als Alternative zu Silizium: Macht der Germanium-Zinn-Transistor dem Quantencomputer Beine?

Empfehlenswert! Diese Neuigkeiten sind schon etwas veraltet, aber dennoch imposant!

"Wissenschaftler des Forschungszentrums Jülich haben einen neuartigen Transistor aus einer Germanium-Zinn-Legierung gefertigt, der gegenüber herkömmlichen Schaltelementen einige Vorteile aufweist. Ladungsträger können sich in dem Material schneller bewegen als in Silizium oder Germanium, was niedrigere Spannungen im Betrieb möglich macht. Der Transistor ist damit ein vielversprechender Kandidat für künftige Low-Power- und High-Performance-Chips und könnte sich als nützlich für die Entwicklung von Quantencomputern erweisen. ..."

Als Alternative zu Silizium: Macht der Germanium-Zinn-Transistor dem Quantencomputer Beine? - ingenieur.de Transistoren aus Silizium stoßen angesichts der rasanten Entwicklung von Quantencomputern oder dem Internet der Dinge langsam an ihre Grenzen. Germanium-Zinn-Transistoren könnten eine vielversprechende Alternative sein.

Jülicher Forschende entwickeln neuen Germanium-Zinn-Transistor als Alternative zu Silizium Wissenschaftler des Forschungszentrums Jülich haben einen neuartigen Transistor aus einer Germanium-Zinn-Legierung gefertigt, der gegenüber herkömmlichen Schaltelementen einige Vorteile aufweist. Ladungsträger können sich in dem Material schneller bewegen als in Silizium oder Germanium, was niedrigere Spannungen im Betrieb möglich macht. Der Transistor ist damit ein vielversprechender Kandidat für künftige Low-Power- und High-Performance-Chips und könnte sich als nützlich für die Entwicklung von Quantencomputern erweisen.

Elektronenmikroskopische Aufnahmen des Germanium-Zinn-Transistors: Der Aufbau folgt einer 3D-Nanodrahtgeometrie, ein Design, das auch für die neueste Generation von Computerprozessoren verwendet wird. (Redaktionelle Verwendung mit Quellenangabe "Forschungszentrum Jülich" erlaubt.)

On the World’s First Universal Processor by Tachyum

I don't remember I have heard of this company before! Tachyum attempts to fuse CPU, GPU and AI accelerator architectures/functionality into one chip. Very exciting!

More competition for Intel and Nvidia is good! Let's see what happens when the dust settles and after the hype goes quieter ...

"... May 2020 ... a new type of processor called Prodigy that was under development by a startup called Tachyum. ...

Tachyum has formally launched Prodigy ... currently using different types of processors to perform different types of tasks — central processing units (CPUs) for general-purpose processing, graphics processing units (GPUs) for graphics and hardware acceleration of algorithms that process large blocks of data in parallel, and AI accelerators for artificial intelligence (AI) applications. ... CPUs do their best work on scalar values, GPUs do their best work on vector values, and AI accelerators do their best work on matrix values. What Prodigy does is to unify the functionality of a CPU, GPU, and TPU into a single architecture that’s implemented on a single monolithic device. ..."

Tachyum Begins Pre-Orders for Prodigy Evaluation Platform "Tachyum™ [6/1/2022] announced that it will be building a limited quantity of Tachyum Prodigy Evaluation Platforms later this year, featuring fully functional Prodigy processors with memory and application software for qualified customers and partners who pre-order between now and July 31, 2022. Tachyum’s Prodigy Evaluation Platform provides a high-performance server in a standard 2U air cooled form factor ..."

Are You Ready to Lay Your Hands on the World’s First Universal Processor? – EEJournal

The company: Tachyum Tachyum is enabling human brain-scale AI and advancing the entire world to a greener era, by delivering the world’s first universal processor.

Credits: https://www.marktechpost.com/

Cerebras' wafer-size chip is 10,000 times faster than a GPU

Wow! Could this be a major breakthrough in chip technology? Wafer scale technology seems to be a very different paradigm in microprocessor technology!

"On a practical level, this means AI neural networks that previously took months to train can now train in minutes on the Cerebras system. ... But Cerebras ... takes that wafer and makes a single, massive chip out of it. Each piece of the chip, dubbed a core, is interconnected in a sophisticated way to other cores. The interconnections are designed to keep all the cores functioning at high speeds so the transistors can work together as one. ... 
The CS-1 beat the Joule Supercomputer [which is No. 82 on a list of the top 500 supercomputers in the world] at a workload for computational fluid dynamics, which simulates the movement of fluids in places such as a carburetor. The Joule Supercomputer costs tens of millions of dollars to build, with 84,000 CPU cores spread over dozens of racks, and it consumes 450 kilowatts of power. In this demo, the Joule Supercomputer used 16,384 cores, and the Cerebras computer was 200 times faster, ... Cerebras costs several million dollars and uses 20 kilowatts of power. ..."

Cerebras' wafer-size chip is 10,000 times faster than a GPU | VentureBeat