Flow Computing raises €4m to boost CPU AI
Flow Computing in Helsinki emerged from stealth with €4m of funding for IP that it says can dramatically boost the performance of any processor core.
Flow Computing is a spin out of the VTT Technical Research Centre in Finland and has developed a licensable parallel processing IP core that can be added to a chip design to boost performance by up to 100x.
The Parallel Processing Unit (PPU) IP can be integrated into any CPU design architecture, instruction set or process geometry and is immediately usable, doubling or tripling performance by optimising the flow of instructions through the core.
- VTT spins out silicon quantum processor business
- VTT spins out MEMS spectrometer startup
- 100x power efficiency boost for AI
The PPU is fully backwards compatible with every existing software application for that CPU architecture. Recompiling the code to make use of the PPU can accelerate applications by as much as 100X, reducing the ned for GPU accelerators in AI applications without any changes to the software.
The more PPU cores integrated on-die, the higher the performance boost that will be subsequently gained and Flow has optimized licenses for every major tier of CPU market application – mobile, PC and supercomputer.
Flow is looking at data centre and AI chip designers as well as developers of chips for autonomous vehicles to provide more AI performance at lower power consumption.
While it boosts the CPU, all other connected units such as matrix units, vector units, NPUs and GPUs also benefit from the performance of PPU and get a major performance increase from the boosted CPU.
Flow is already in initial discussions with major semiconductor vendors from around the world and more technical details will be publicly shared during the second half of 2024.
The €4m pre-seed round funding comes from Butterfly Ventures (lead VC), FOV Ventures, Sarsia, Stephen Industries, Superhero Capital and Business Finland. VTT maintains an equity stake in Flow Computing and all the patented IP developed over several years during this period has been officially transferred to Flow.
“We firmly believe there have been only incremental improvements in CPU performance during the last few decades – in our opinion, this has led to a situation where the CPU has actually become the weakest link in computing due to its sub-optimal sequential architecture,” said Timo Valtonen, co-founder and CEO of Flow Computing.
“A new era in CPU performance has become a necessity in order to meet the continuously increasing demand for more computing performance, driven to large extent by needs in AI, as well as edge and cloud computing. Flow intends to lead the SuperCPU revolution through its radical new Parallel Performance Unit (PPU) architecture, enabling up to 100X the performance of any CPU, regardless of architecture and with full backwards software compatibility.”
“At VTT we are excited by the technical innovations of Flow Computing as Finland moves to become a world-leading exporter of cutting-edge semiconductor designs. They are providing a complete performance ecosystem to not only the CPU vendors, but the development communities that have coalesced around major CPU designs such as X86, Apple M-Series, Exynos, ARM and RISC-V. We are excited that Flow’s technology which has been developed at VTT is able to contribute in a significant way to the not just to Finnish and European computing ecosystem, but in the future of the whole computing industry!” said Antti Vasara, CEO of VTT.
“Every industry in tech continues to suffer due to the fact that the rate of CPU improvements has slowed over the last decade. Flow is the first major departure from this trend, offering multiples of performance instead of some percentage points and – in our opinion – Flow will have a broader impact on the baseline performance of the computing market than for example quantum computing in this decade. Despite the huge investments in AI, general purpose computing will dominate its cost and constrain its capability. Flow Computing is solving this problem by enabling the next generation of SuperCPUs to easily surpass current industry leaders such as the Apple M-series, Nvidia Grace, Google Axion and Microsoft Azure Cobalt 100,“ said Juho Risku, Partner & co-founder of Butterfly Ventures.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
