Utililizing Texture Cache in SYCL Applications Internship
SYCL is a C++-based heterogeneous programming standard for accelerator computing. Codeplay works on implementations of this standard and are investigating creating and using extensions to the language to allow applications to fully use target hardware. This includes extensions to take advantage of GPU texture caches for efficient random access caching being implemented in the implementation of SYCL on top of CUDA. This is particularly useful in Molecular Dynamics applications but also some machine learning algorithms.
The original purpose of the texture cache in CUDA was to optimize read-only accesses of image data, and this cache can be most efficiently used by taking account of the 2D/3D spatial arrangement of that data. Correspondingly in 3D molecular dynamics simulations the indexing of molecules in data arrays can take account of the texture cache to boost the performance of read-only access patterns that are used by standard algorithms. In SYCL this can be accessed via an extension (https://github.com/intel/llvm/pull/7397). As a SYCL extension, however, use of the extension should be integrated in a way that provides a clean fallback if the extension is not present. Thus this internship will involve using this extension in a few existing frameworks and molecular dynamic applications, determining how much gain the extension use provides, and ensuring that performance is not impacted on hardware that does not provide the same texture cache benefits. As a stretch goal it could be possible to investigate how to add to the extension such that the programmer can access "hint" instructions to suggest where future accesses will be.
The ideal candidate will be someone who has a strong interest in heterogeneous programming, who is able to progress on tasks autonomously while sharing progress or blockers with their team.
This internship is on offer for a duration of 3-6 months. Although we usually offer internships for the summer, we are delighted to accept applications for longer or shorter internships and many of our interns have stayed with us to do university placement years or have become permanent members of our team.
The position is temporary and will be based at our head office in central Edinburgh. The successful candidate must, by the start of their employment, have successfully obtained the right to work in the UK.
Starting salary - Competitive & Negotiable
- Work as part of a team implementing SYCL on various GPUs and supporting HPC applications
- Extend the Kokkos framework to take advantage of a SYCL extension for GPU texture cache loads
- Improve you knowledge about Nvidia hardware with a focus on the texture cache
- Work on state-of-the-art, production level Molecular Dynamics projects
- Gain experience with optimizing and benchmarking heterogeneous codebases
- Improve your knowledge of C++, SYCL, and heterogeneous programming
What We Offer
Codeplay is internationally recognized as being at the forefront of software development technology. Working at Codeplay offers unique opportunities and challenges to learn new technologies and to get involved in shaping the software industry. Codeplay's projects range from: creating tools and tool sets (optimizing compilers, debuggers, profilers, programming models & APIs, test systems/suites, etc.) for brand-new unannounced next-generation hardware; developing and optimizing current machine learning (TensorFlow™) and computer vision applications; research and development into lowering the power consumption of mobile devices; and contributing to new programming models and software standards (e.g. C++, SYCL™, OpenCL™, Vulkan® and SPIR™-V).v
Codeplay Software is a world pioneer in enabling acceleration technologies used in AI, HPC and automotive. Codeplay was established in 2002 in Edinburgh, Scotland and developed some of the first tools enabling complex software to be accelerated using graphics processors. Today, most AI software is developed using graphics processors designed for video games, and more recently specialised AI and computer vision accelerators. Codeplay continues to work with global technology leaders to make the latest complex AI systems programmable using open-standards based programming languages and allows application developers to quickly bring software to the market. Codeplay is also deeply involved with the definition of open standards, especially OpenCL™, SPIR™, SYCL™, and Vulkan™ through The Khronos Group, and MISRA C++ for automotive.
- A friendly and relaxed working environment where colleagues respect and support each other.
- Learn from experienced colleagues: our staff frequently give company-wide knowledge-sharing presentations on new software and technology trends, personal projects and other relevant subjects.
- Great career prospects within a growing company
- Newstart mentoring scheme
- Flexible working hours
- Frequent company social events (social nights, pot luck lunches, gaming, karting, outdoor adventures, and more)
The ideal candidate is a self-motivated individual who can work through a problem from start to finish, meet deadlines as required to support software projects, and has the ability to work under pressure. You will be working on technology that is seldom announced to the public and could be several years away from public release, so confidentiality is critical. You should be able to work both independently and collaboratively within teams of varying skill sets and expertise. You should be welcoming towards, and able to adapt to, the introduction of new technologies, responsibilities and projects, as Codeplay works with cutting-edge technology and takes on a diverse range of projects.
- C++ development experience
- Some level of GPU programming experience, e.g. CUDA, OpenCL, or SYCL.
- Good written English and communication skills
- Working knowledge of source control and configuration management tools
Knowledge and/or experience of any of the following would be useful for the role, but are not essential. You must be able to learn new skills, but will be given support from Codeplay's team to get you up to speed.
- Knowledge of GPU programming optimization techniques.
- Experience with benchmarking programs.
- Experience with HPC programming and applications.
- Some familiarity with Molecular Dynamics algorithms and/or Machine Learning algorithms and frameworks.
Any application which results in a job offer is conditional upon the candidate already having or successfully applying for the right to work in the UK. Codeplay is proud to be an Equal Opportunities employer. We celebrate and value diversity. We are committed to creating an inclusive environment for all employees, regardless of race, colour, gender, religion, national origin, ancestry, citizenship, disability, age, sexual orientation or any other characteristic protected by law. All employment is decided on the basis of qualifications, merit and business need and we strongly encourage applicants from all backgrounds and walks of life. If you have a disability or additional need that requires accommodation, just let us know.
The Cover Note/Letter is just as important as your CV/Resume, so please include one, stating why you believe you are a good fit for the position. If you have examples of relevant work or projects, we would love to see them, so please include links to them in your application.You should be studying towards a degree in a relevant subject; however, if you can prove skills and knowledge through portfolio work, projects or previous employment, then a degree is not essential.If you have C++ Code Samples (where relevant) please put the link to these on your application; Our system will only accept links to personal projects/GitHub accounts or PDF documents.
We aim to close this advert on the 28th April 2023, so please look
to get your application in prior to this date. We anticipate a high
level of applications for this role and so reserve the right to close
this vacancy prior to the closing date given.