Graphcore
 | |
| Company type | Private |
|---|---|
| Industry | Semiconductors |
| Founded | 2016  |
| Headquarters | , |
Key people | Nigel Toon (CEO) Simon Knowles (CTO) |
| Products | IPU, Poplar |
| Website | https://www.graphcore.ai/ |
Graphcore is a British semiconductor company that develops accelerators for AI and machine learning. It aims to make a massively parallel Intelligence Processing Unit (IPU) that holds the complete machine learning model inside the processor.[1]
History
Graphcore was founded in 2016 by Simon Knowles and Nigel Toon.
In the autumn of 2016, Graphcore secured a first funding round lead by Robert Bosch Venture Capital. Other backers include Samsung, Amadeus Capital Partners, C4 Ventures, Draper Esprit, Foundation Capital, and Pitango.[2][3]
In July 2017, Graphcore secured a round B funding lead by Atomico,[4] which was followed a few months later by $50 million in funding from Sequoia Capital.[5]
In December 2018, Graphcore closed its series D with $200 million raised at a $1.7 billion valuation, making the company a unicorn. Investors included Microsoft, Samsung and Dell Technologies.[6]
On 13 November 2019, Graphcore announced that their Graphcore C2 IPUs are available for preview on Microsoft Azure.[7]
Products
In 2016, Graphcore announced the world's first graph tool chain designed for machine intelligence called Poplar Software Stack.[8][9][10]
In July 2017, Graphcore announced their first chip, called the Colossus GC2, a "16 nm massively parallel, mixed-precision floating point processor", first available in 2018.[11][12] Packaged with two chips on a single PCI Express card called the Graphcore C2 IPU, it is stated to perform the same role as a GPU in conjunction with standard machine learning frameworks such as TensorFlow.[11] The device relies on scratchpad memory for its performance rather than traditional cache hierarchies.[13]
In July 2020, Graphcore presented hardware using a second generation processor called GC200 built in TSMC's 7nm FinFET manufacturing process. GC200 is a 59 billion transistor, 823 square-millimeter integrated circuit with 1,472 computational cores and 900 Mbyte of local memories.[14]
Both the older and newer chips can use 6 threads per tile (for 8,832 threads in total, per GC200 chip) "MIMD (Multiple Instruction, Multiple Data) parallelism and has distributed, local memory as its only form of memory on the device" (except for registers), and the newer GC200 chip has about 630 KiB per tile, vs 256 KiB per tile in older C2 chip, that are arranged into islands (4 tiles per island),[15] that are arranged into columns, and latency is best within tile. The IPU uses IEEE FP16, with stochastic rounding, and also single-precision FP32, at lower performance.[16] Code and data executed locally must fit in a tile, but with message-passing, all on-chip or off-chip memory can be used, and software for AI makes it transparently possible, e.g. has PyTorch support.
See also
References
- ^ Peter Clarke (2016-11-01). "AI Chip Startup Shares Insights: "Very large" FinFET chip in the works at TSMC". eetimes. Archived from the original on 2017-08-03. Retrieved 2017-08-02.
- ^ Arjun Kharpal (2016-10-31). "AI chipmaker Graphcore raises $30 million to take on Intel". CNBC. Archived from the original on 2017-07-31. Retrieved 2017-07-31.
- ^ Madhumita Murgia (2016-10-31). "UK chip start-up Graphcore raises £30m for take on AI giants". Financial Times. Archived from the original on 2017-08-03. Retrieved 2017-08-02.
- ^ Jeremy Kahn and Ian King (2017-07-20). "U.K. Chip Designer Graphcore Gets $30 Million to Fund Expansion". Bloomberg. Archived from the original on 2017-07-31. Retrieved 2017-07-31.
- ^ Lynley, Matthew (2017-11-12). "Graphcore raises $50M amid a flurry of AI chip activity". TechCrunch. Archived from the original on 2017-12-08. Retrieved 2017-12-07.
- ^ "AI chip startup Graphcore closes $200M Series D, adds BMW and Microsoft as strategic investors". TechCrunch. Archived from the original on 2018-12-18. Retrieved 2018-12-19.
- ^ Toon, Nigel. "Microsoft and Graphcore collaborate to accelerate Artificial Intelligence". www.graphcore.ai. Archived from the original on 2019-11-16. Retrieved 2019-11-16.
- ^ Fyles, Matt. "Inside an AI 'brain' - What does machine learning look like?". www.graphcore.ai. Archived from the original on 2019-11-16. Retrieved 2019-11-16.
- ^ Doherty, Sally. "Introducing Poplar® - our IPU-Processor software at NeurIPS". www.graphcore.ai. Archived from the original on 2019-11-16. Retrieved 2019-11-16.
- ^ Fyles, Matt. "Graph computing for machine intelligence with Poplar™". www.graphcore.ai. Archived from the original on 2019-11-16. Retrieved 2019-11-16.
- ^ a b Trader, Tiffany (2017-07-20). "Graphcore Readies Launch of 16nm Colossus-IPU Chip". hpcwire.com. HPC Wire. Archived from the original on 2017-12-12. Retrieved 2017-12-11.
- ^ Lucchesi, Ray (2018-11-19). "New GraphCore GC2 chips with 2PFlop performance in a Dell Server". silvertonconsulting.com. Silverton Consulting. Archived from the original on 2018-12-17. Retrieved 2018-12-16.
- ^ Citadel High Performance Computing R&D Team (2019). "Dissecting the Graphcore IPU Architecture via Microbenchmarking" (PDF). Archived (PDF) from the original on 2020-06-21. Retrieved 2020-06-29.
- ^ "Graphcore Introducing 2nd Generation IPU Systems For AI At Scale". Archived from the original on 2020-08-04. Retrieved 2020-08-09.
- ^ Jia, Zhe; Tillman, Blake; Maggioni, Marco; Daniele Paolo Scarpazza (2019). "Dissecting theGraphcore IPUArchitecturevia Microbenchmarking" (PDF). arXiv:1912.03413. Archived (PDF) from the original on 2021-10-09. Retrieved 2022-01-21.
- ^ "THE GRAPHCORE SECOND GENERATION IPU" (PDF). Archived (PDF) from the original on 2022-01-21. Retrieved 2022-01-21.