Collaboration brings direct benefits to healthcare company

The project was part of the CodeOpt Scotland programme, which was designed to increase company competitiveness by optimising the performance of business-critical software. CodeOpt was run in partnership with EPCC, Scottish Enterprise and Intel.

The project with Optos set out to identify a suitably priced Intel multicore processor on which to run the company's code, and to accelerate an important Optos machine-learning code. The project baseline was Optos code running on a specific Intel chip and the target was to reduce its runtime to below 12 seconds on the preferred chip.

Turning academic excellence into commercial benefits

EPCC achieved a runtime of 6.6 seconds on the preferred Intel chip; that’s almost two times faster than the target, over 15 times faster than the baseline software-hardware combination, and five times faster than the baseline software running on the new chip.

Importantly for Optos, the preferred Intel chip is 2.5 time faster than the original baseline when running the optimised codes, while being 25% cheaper than the target price.

In addition, EPCC accelerated a second key Optos code to run in just over five minutes on the preferred hardware, achieving the same acceleration figures as for the main code.

Finally, EPCC’s investigations into the popular TensorFlow machine learning framework allowed Intel to improve developer documentation, especially for the Windows platform, and led to an acceleration of the default, open source version of the Intel MKL-DNN library to the benefit of the wider community.

“The success of this project highlights EPCC’s capability to translate academic excellence to long-lasting benefit for our commercial partners”. Prof. Mark Parsons, Director, EPCC

Business advantages

The Optos codes optimised by EPCC with Intel’s help are now fast enough to be deployed on a production device. Additionally, we identified an Intel chip that is cheaper than the previous candidates and is better suited to Optos’ codes. This combination allows the company to cut costs while enhancing the performance of its product and will further boost Optos’ market share as a whole, helping promote sales of its other related products.

Optos has gained enhanced insight into the nature and the bottlenecks of its code through this project, and is now equipped with a number of ideas to further improve it. The company also has better understanding of advanced techniques to benchmark and optimise multi-core codes, as used by EPCC for this project, and will apply this new knowledge to its software for years to come.

“This project has brought direct benefits to our business, both by enabling us to develop competitive products faster and by building essential knowledge within our organisation. We look forward to a continued productive relationship with EPCC.” Dr Jano van Hemert, Research Director, Optos

Machine-learning advances

Intel gained insights into the performance of the MKL-DNN mathematics library employed by TensorFlow. In the seven months that it was active, the project established that MKL-DNN has been accelerated by up to 4.5 times in our use-cases, information which is important for the uptake of the library. EPCC also provided profiling information gleaned from the Optos use-cases and this will help Intel further improve the performance of MKL-DNN.

Intel improved Windows documentation in particular, which will increase the uptake of TensorFlow and MKL-DNN.

“We are excited to partner with Optos to identify key computing capabilities needed to accelerate their critical machine-learning codes. Intel processors demonstrated a significant boost in the performance for their codes, outperforming the target time of the currently deployed technology.” Trish Damkroger, Vice President and General Manager, Intel’s Data Center Group

Wider benefits for the partners and the country

It is expected that this new opportunity will support ongoing product development activities at Optos' headquarters in Dunfermline, supporting jobs in areas from research and development through to manufacturing. This will maintain high-value employment in Fife, as it continues to enhance Scotland’s reputation in the areas of retinal imaging and machine learning.

EPCC increased its exposure to machine-learning codes as utilised in cutting-edge industrial applications and developed in-depth understanding of the industry-leading TensorFlow library. We will exploit this new knowledge in future projects with our industrial and academic partners, notably in our leading role in the Edinburgh and South East Scotland City Regional Deal.

“The Optos CodeOptScotland project showcases the impact of engaging Intel for the benefit of Scotland’s economy, Scotland-based companies and Scottish academia.” David Smith, Director of Technology, Scottish Enterprise 

Further information

To explore how EPCC can work with your organisation, contact our Business Team.

EPCC's recent and current collaborative research projects.

Intel

Optos plc

Scottish Enterprise