Published October 8, 2020 | Version 1.0
Thesis Open

Adding CUDA® Support to Cling: JIT Compile to GPUs

  • 1. Helmholtz-Zentrum Dresden-Rossendorf
  • 2. Helmholtz-Zentrum Dresden-Rossendorf, Lawrence Berkeley National Laboratory
  • 3. CERN

Description

Jupyter Notebooks are omnipresent in the modern scientist's and engineer's toolbox just as CUDA C++ is in accelerated computing. We present the first implementation of a CUDA C++ enabled read-eval-print-loop (REPL) that allows to interactively "script" the popular CUDA C++ runtime syntax in Notebooks. With our novel implementation, based on LLVM, Clang and CERN's C++ interpreter Cling, the modern CUDA C++ developer can work as interactively and productively as (I)Python developers while keeping all the benefits of the vast C++ computing and library ecosystem coupled with first-class performance.

Knowledge in developing applications with GPU acceleration is recommended for the talk.

Files

Cling-CUDA_SimeonEhrig_Final.mp4

Files (230.3 MB)

Name Size Download all
md5:66e498df0c18c5dfe55ce6d26098371c
228.9 MB Preview Download
md5:91fff7803a4b1ad56c6ff26cb50af6fd
89.0 kB Preview Download
md5:ebf5c7a7be77d3544cca535051aeab46
576.0 kB Preview Download
md5:493518fe653c1fec9b4af87c7a455224
661.5 kB Preview Download