Roku Engineering Symposium: Modern C++ and GPU accelerated applications

Roku Denmark will again this year host a Symposium with exciting speakers. This year we have the pleasure of a visit from Bryce Adelstein Lelbach from NVIDIA. A bit about Bryce:

Bryce Adelstein Lelbach has spent over a decade developing programming languages, compilers, and software libraries. He is passionate about parallel programming and strives to make it more accessible for everyone.

Bryce is a Principal Architect at NVIDIA, where he leads programming language efforts and drives the technical roadmap for NVIDIA's compute compilers and libraries.

He is one of the leaders of the systems programming language community, having served as chair of the Standard C++ Library Evolution group and the US standards committee for programming languages (INCITS/PL22). He has been an organizer and program chair for many conferences over the years. He previously worked at Lawrence Berkeley National Laboratory and Louisiana State University. The event will be in English.

About Roku: Founded by Anthony Wood, Roku pioneered streaming to the TV. Today, Roku streaming devices are used by millions of consumers in North America, Latin America, and parts of Europe. Roku is the No. 1 TV streaming platform in the U.S., Canada, and Mexico by hours streamed. Roku makes TV's, streaming players, wired and wireless audio products, and recently also started a Smart Home (IOT) portfolio. Roku Denmark contributes to many of Roku's products - both on software, hardware and acoustics. The embedded software in Roku's products is mainly C++ and many of our colleagues are strong C++ programmers.

Below are brief abstracts for the presentations:

Bryce will be presenting the following:

Getting the most out of your GPU with C++ doesn't require writing custom kernels or manually managing storage for everything! Come learn about the libraries and techniques that make writing CUDA C++ code easier and more performant. Through examples, we'll explore all aspects of writing modern C++ software for GPUs, including heterogeneous memory management, algorithm design, and synchronization.

During this talk, you'll:

· Learn to evaluate when you should use a CUDA library versus writing your own kernel.

· Explore popular CUDA C++ libraries such as Thrust, CUB, and libcu++.

· Understand how you can easily compose different CUDA libraries and your own custom CUDA C++ code together.

· Build intuition about the performance implications of CUDA libraries.

· You'll leave confident about how to select the best tool for the job to accelerate your C++ applications for your unique use cases.

From Roku we will this year have Jonas Hansen and Andreas Hermann presenting the following:

Size matters, in embedded software. That is, on embedded platforms the size of the generated code and the size of any compile time data is important to minimize. Moreover, keeping the amount of dynamic relocations at a minimum saves on load-time and memory usage. To this end, Roku has developed data containers to store compile time data in an compact format. Here compact refers to the size of the code, data, and the number of dynamic relocations. Moreover, since all the data is known at compile time the data containers are able to figure out how to optimally access the data, be it via binary search, linear search, hashing, etc.

Location:

• Auditorium, Roku Denmark, Hedeager 44, DK-8200 Aarhus N
• Date: Thursday 18th September, 15:00

• 15.00 Opening and welcome - Jasper Knudsen (Roku)    
• 15.05 Size matters, in embedded SW - Jonas Hansen, Andreas Hermann (Roku) 
• 15:50 Break 
• 16.05 Modern C++ for GPU’s - Bryce Adelstein Lelbach (NVIDIA) 
• 17.30 Snacks and demo's in the basement
• 18:00 End of the event