Jacob Tomlinson | Tech Exeter Conference 2020 Intro to GPU development in Python 

No content

3 RAPIDS Github https://github.com/rapidsai 

4 GPU-Accelerated ETL The Average Data Scientist Spends 90+% of Their Time in ETL as Opposed to Training Models 

5 Lightning-fast performance on real-world use cases Up to 350x faster queries; Hours to Seconds! TPCx-BB is a data science benchmark consisting of 30 end-to-end queries representing real-world ETL and Machine Learning workflows, involving both structured and unstructured data. It can be run at multiple “Scale Factors”. ▸ SF1 - 1GB ▸ SF1K - 1 TB ▸ SF10K - 10 TB RAPIDS results at SF1K (2 DGX A100s) and SF10K (16 DGX A100s) show GPUs provide dramatic cost and time-savings for small scale and large-scale data analytics problems ▸ SF1K 37.1x average speed-up ▸ SF10K 19.5x average speed-up (7x Normalized for Cost) 

6 Dask DEPLOYABLE ▸ HPC: SLURM, PBS, LSF, SGE ▸ Cloud: Kubernetes ▸ Hadoop/Spark: Yarn PYDATA NATIVE ▸ Easy Migration: Built on top of NumPy, Pandas Scikit-Learn, etc ▸ Easy Training: With the same APIs ▸ Trusted: With the same developer community EASY SCALABILITY ▸ Easy to install and use on a laptop ▸ Scales out to thousand node clusters POPULAR ▸ Most Common parallelism framework today in the PyData and SciPy community 

7 What are GPUs? 

8 Gaming Hardware For pwning n00bs 

9 Mysterious Machine Learning Hardware For things like GauGAN Semantic Image Synthesis with Spatially-Adaptive Normalization Taesung Park, Ming-Yu Liu, Ting-Chun Wang, Jun-Yan Zhu arXiv:1903.07291 [cs.CV] 

10 CPU GPU 

11 https://youtu.be/-P28LKWTzrI 

12 GPU vs CPU https://docs.nvidia.com/cuda/cuda-c-programming-guide/ 

13 Using a GPU is like using two computers Icons made by Freepik from Flaticon Network SSH VNC RD SCP FTP SFTP Robocopy 

14 Using a GPU is like using two computers Icons made by Freepik from Flaticon PCI CUDA 

15 What is CUDA? 

16 CUDA 

17 I don’t write C/C++ 

18 What does CUDA do? Construct GPU code with CUDA C/C++ language extensions Copy data from RAM to GPU Copy compiled code to GPU Execute code Copy data from GPU to RAM How do we run stuff on the GPU? 

19 Let’s do it in Python 

20 

21 

22 Writing a Kernel Differences between a kernel and a function ● A kernel cannot return anything, it must instead modify memory ● A kernel must specify its thread hierarchy (threads and blocks) A kernel is a GPU function 

23 Threads, blocks, grids and warps https://docs.nvidia.com/cuda/cuda-c-programming-guide/ 

24 What? Rules of thumb for threads per block: ● Should be a round multiple of the warp size (32) ● A good place to start is 128-512 but benchmarking is required to determine the optimal value. 

25 Imports 

26 Data arrays 

27 Example kernel 

28 Running the kernel 

29 Absolute positions 

30 Thread and block positions 

31 Example kernel (again) 

32 How did the GPU update our numpy array? If you call a Numba CUDA kernel with data that isn’t on the GPU it will be copied to the GPU before running the kernel and copied back after. This isn’t always ideal as copying data can waste time. 

33 Create a GPU array 

34 Simplified position kernel 

35 GPU Array 

36 Copy to host 

37 Higher level APIs 

38 

39 cuDF 

40 cuDF cuIO Analytics Data Preparation Visualization Model Training cuML Machine Learning cuGraph Graph Analytics PyTorch, TensorFlow, MxNet Deep Learning cuxfilter, pyViz, plotly Visualization Dask GPU Memory RAPIDS End-to-End GPU Accelerated Data Science 

41 Interoperability for the Win DLPack and __cuda_array_interface__ mpi4py 

42 __cuda_array_interface__ 

43 Recap 

44 Recap GPUs run the same function (kernel) many times in parallel When being called each function gets a unique index CUDA/C++ is used to write kernels, but high level languages like Python can also compile to it Memory must be copied between the CPU (host) and GPU (device) Many familiar Python APIs have GPU accelerated implementations to abstract all this away Takeaways on GPU computing 

THANK YOU Jacob Tomlinson @_jacobtomlinson