What we are? We are a team of professional designers of multicore processors.

 

We started our project in 2011. At that time, we claimed that in 5-10 years there would be hundreds of computing cores in a processor, energy efficiency would be more significant than productivity, productivity would be achievable, thanks to the specialization of computing units and parallel processing, and microarchitecture would be determined by energetics and technology. We set the goal - to become designers of manycore processors at world level by 2020. We gathered a strong team and today we’re a working fabless design center. Our processors are manufactured by the largest factories using wide range of technologies up to 16 nm. We create complete program solutions and design not only the chips but all the necessary software infrastructure.

 more info…

 

Our product: MALT - truly manycore processors

 

MALT (Manycore Architecture with Lightweight Threads) - specialized energy-efficient processors with hundreds of extremely compact computing cores on a single chip. MALT processor family has models with scalar, vector and combined architectures. Solutions based on MALT show the higher performance-per-watt than customized FPGA systems, surpassing them by performance-per-dollar value. MALT architecture programming complexity is on a par with the universal x86/GPU/ARM computer systems and they show the performance-per-watt of a much higher order.

more info…

 

MALT applications? Our processors are relevant for energy-efficient solutions, blockchain, big data and mathematical physics

 

Are you developing the systems based on blockchain technology, building trusted environment, processing big data or solve complex physical and mathematical problems? In any case, your software is implemented by ‘some’ hardware where the productivity and energy-efficiency are required. A universal processor handles it, but it will definitely lose to the specially customized for your tasks computer by 1-2 orders of magnitude. The MALT processor performance-per-watt is as high as one of modern FPGA systems surpassing them many times by performance-per-dollar value.

more info...

 

How and why does it work?

 

Single-type computing elements produced ‘from scratch’ for a specific class of tasks are approximately 80% of any chip we develop. These elements determine productivity and chip power. There are hundreds to several thousands elements on the chip! All the computing elements are programmed in C language or in its subset. The elements are combined into groups controlled by compact universal RISC processors grouped into computational array that to a programmer looks like a typical multi-threaded processor programmed in C/C++ . he classic problems of multithreaded processing (conflict resolution mechanism of simultaneous access to shared data including atomic operations) are solved in hardware at the memory controller level.

more info...

News

2012—2024 © All rights reserved.