REPLICA Multiprocessor
Revolutionary multiprocessor framework for efficient parallel computing. With respect to current CPUs, provides processors with radically higher performance and better productivity in SW engineering. REPLICA uses parallelism as a resource for better performance, not a productivity bog.
REPLICA team has developed a revolutionary new multiprocessor framework and architecture. It defines a new way how to build processors for application-specific embedded systems, cloud computing as well as general-purpose devices that require ultimate performance. REPLICA is a family of patented and parametric parallel architectures enabling constellations optimized for different use cases and performance levels. Multiprocessors designed according to the principles of the framework treat parallelism as a natural design factor and a resource for higher performance rather than an obstacle that decreases productivity and causes hard-to-debug performance issues. Through its unique combination of architectural and methodological techniques and solutions, REPLICA helps to eliminate the performance and programmability bottlenecks of current multicore processors and fundamental hurdles that have prevented using the benefits of parallel computing in the mainstream of software engineering.
REPLICA provides backwards compatibility with the existing software base, tools and system software.
Our experimentations indicate from tenfold to millionfold performance increase combined with 5 X software development productivity jump for parallel functionalities.
Recently we received our first financial investment to boost the commercialization of our technology. Currently we are contacting companies interested in trialing out our technology as well as potential investors for the upcoming seed round in 2022.
Find out more about REPLICA and contact us!
What we are looking for
REPLICA team is looking for funding, especially a lead investor with understanding and contact networks of the microprocessor industry / market segment. Alternatively, there is a look for processor companies to collaborate with.
Meet our team
Main architect: Processor, interconnect and memory system architecture, programming methodology, optimizations, simulation, benchmarking and the theory behind the technology.
Martti.Forsell@vtt.fi
+358 40 581 0249
HW designer: HDL modeling, synthesis, implementation and testing.
Jussi.Roivainen@vtt.fi
+358 40 861 5096