The battle over Heterogeneous Computing

In the Haifa::C++ June event, Oren Ben Simhon will talk about the potential and challenges involved with heterogeneous computing, which includes accelerators such as the GPU, DSP processors, FPGAs, etc. While such processors are applicable to many C++-related use cases, the language itself doesn't support them. Fortunately, there are various extensions and libraries that can help. June 17, 17:30, at Mobileye's office in Matam.

Agenda:
17:30 - Gathering, mingling, refreshments consumption
18:00 - The battle over Heterogeneous Computing, by Oren Ben Simhon
19:00 - Wrap up, mingling, refreshments annihilation

Abstract:
As our world advances to AI and dense computing, more and more GPUs and DSPs are developed to accelerate computation. But how can one send C++ kernels to be executed on a GPU? Unfortunately, C++ does not yet propose a solution to heterogeneous systems, but it is evolving, and multiple solutions are on their way.
In the lecture, Oren will describe existing solutions like OpenCL, SYCL, CUDA, new C++26 libraries, and other programming languages.
Further, he will present the compiler technology that was developed in Mobileye based on LLVM infrastructure and the challenges they have faced.

Bio:
Oren Ben Simhon is a seasoned compiler engineering director at Mobileye SW, spearheading R&D efforts to develop cutting-edge compilers and runtime libraries for Mobileye's EyeQ system on chip.
A graduate of the Technion with a BSc in computer engineering and an MBA, Oren brings over 18 years of expertise to the field of embedded systems programming and design. His academic background, including former TA in the theory of compilation course at the Technion, coupled with his extensive industry experience, has solidified his reputation as a thought leader in compiler technology.

Our kind hosts and the event sponsors:
Mobileye is leading the mobility revolution with its autonomous-driving and driver-assist technologies, harnessing world-renowned expertise in computer vision, machine learning, mapping, and data analysis. Our technology enables self-driving vehicles and mobility solutions, powers industry-leading advanced driver assistance systems, and delivers valuable intelligence to optimize mobility infrastructure.