The Silicon Review
“We’re optimistic for a future where people live healthier, more informed, more creative lives”
Graphcore has designed a new processor specifically for machine learning - the intelligence processing unit (IPU) to let innovators create next generation machine intelligence. We’re always interested in hearing from talented software, hardware and silicon engineers to join our team. We see a world where technology enhances human potential, and takes us into a new era of intelligence and progress that everyone can benefit from. Graphcore was earlier this year named one of the most disruptive startups in the world.
The investors of the company are Amadeus Capital, Robert Bosch Venture Capital, C4 Ventures, Dell Technologies Capital, Draper Esprit, Foundation Capital, Pitango and Samsung Catalyst Fund. ARM founder Hermann Hauser is also an early investor in the firm. Google DeepMind founder Demis Hassabis and Zoubin Ghahramani, Uber’s chief scientist and Cambridge university professor, are among new investors in Graphcore, which makes chips that can process machine learning faster and more efficiently. It comes hot on the heels of a $30m (£23m) raise only in October last year. While the latest cash was not a necessity, the calibre of interested investors after the oversubscribed series A led Graphcore to take on the new round of investment, which was led by VC firm Atomico.
Our IPU accelerators and Poplar software framework together make the fastest and most flexible platform for current and future machine intelligence applications, lowering the cost of AI in the cloud and datacenter, improving performance and efficiency by between 10x to 100x.
Graphcore systems excel at both training and inference. The highly parallel computational resources together with graph software tools and libraries, allows researchers to explore machine intelligence across a much broader front than current solutions. This technology lets recent success in deep learning evolve rapidly towards useful, general artificial intelligence.
Intelligence Processing Unit - IPU
We’ve created a completely new processor that’s the first to be specifically designed for machine intelligence workloads – an Intelligence Processing Unit (IPU) that will set a new pace of innovation. The IPU has been optimized to work efficiently on the extremely complex high-dimensional models needed for machine intelligence workloads. It emphasizes massively parallel, low-precision floating-point compute and provides much higher compute density than other solutions.
Our IPU systems are designed to lower the cost of accelerating AI applications in cloud and enterprise datacenters to increase the performance of both training and inference by up to 100x compared to the fastest systems today.
Poplar- Graph programming framework
Poplar is our graph programming framework for IPU accelerated platforms, designed to meet the needs of customers in advanced research teams and those working on commercial deployments in a wide range of industry sectors. It’s a C++ framework that provides a seamless interface to standard machine learning frameworks, like Tensorflow and MXNet, so your existing applications written for Tensorflow will work out of the box on an IPU.
To support this, Poplar has a full set of drivers, application libraries and debugging and analysis tools to help tune performance and a C++ and Python interface for application development making our IPU systems quick and easy to use. The Poplar framework is completely scaleable; it is a natural choice for all machine intelligence programming paradigms.
The Founding Team
Nigel Toon, CEO: Nigel is co-founder & CEO of Graphcore. Nigel was CEO of two VC-backed silicon companies before founding Graphcore; Picochip, which was sold to Mindspeed in 2012 and most recently, XMOS, in which Graphcore was incubated for two years before being established as a separate ] entity in 2016. Before that he was co-founder of Icera, a 3G cellular modem chip company, where he led ] Sales and Marketing and was on the Board of Directors. Icera was sold to NVIDIA in 2011 for $435M. Prior to Icera, he was Vice President and General Manager at Altera Corporation where he spent 13 years and was responsible for establishing and building the European business unit that grew to over $400m annual revenues. He is the author of 3 patents, is chairman of the board of directors at XMOS Ltd and a non-executive director at Imagination Technologies PLC.
Simon Knowles, CTO: Simon is co-founder & CTO of Graphcore. Simon has a strong track record as both engineer and entrepreneur, having co-founded and exited two highly successful fabless semiconductor companies, Element14 and Icera, for a combined value of over $1billion. Before that he headed the microprocessor development group at ST Micro. He is well known as a leading microprocessor designer, responsible for ground-breaking designs at STMicro, Element14, and Icera. He holds an MA in Electrical Science from Cambridge University and is the author of 14 issued patents.