Boost.Compute GTC 2015
- 1. Boost.Compute Kyle Lutz A C++ library for GPU computing
- 2. “STL for Parallel Devices” GPUs (NVIDIA, AMD, Intel) Multi-core CPUs (Intel, AMD) FPGAs (Altera, Xilinx) Accelerators (Xeon Phi, Adapteva Epiphany)
- 4. Iterators buffer_iterator<T> constant_buffer_iterator<T> constant_iterator<T> counting_iterator<T> discard_iterator function_input_iterator<Function> permutation_iterator<Elem, Index> transform_iterator<Iter, Function> zip_iterator<IterTuple> array<T, N> dynamic_bitset<T> flat_map<Key, T> flat_set<T> stack<T> string valarray<T> vector<T> Containers bernoulli_distribution default_random_engine discrete_distribution linear_congruential_engine mersenne_twister_engine normal_distribution uniform_int_distribution uniform_real_distribution Random Number Generators
- 5. Library Architecture OpenCL GPU CPU FPGA Boost.Compute Core STL-like API Lambda Expressions RNGs Interoperability
- 6. Why OpenCL? (or why not CUDA/Thrust/Bolt/SYCL/OpenACC/OpenMP/C++AMP?) • Standard C++ (no special compiler or compiler extensions) • Library-based solution (no special build-system integration) • Vendor-neutral, open-standard
- 8. Low-level API • Provides classes to wrap OpenCL objects such as buffer, context, program, and command_queue. • Takes care of reference counting and error checking • Also provides utility functions for handling error codes or setting up the default device
- 9. Low-level API #include <boost/compute/core.hpp> // lookup default compute device auto gpu = boost::compute::system::default_device(); // create opencl context for the device auto ctx = boost::compute::context(gpu); // create command queue for the device auto queue = boost::compute::command_queue(ctx, gpu); // print device name std::cout << “device = “ << gpu.name() << std::endl;
- 11. High-level API
- 12. Sort Host Data #include <vector> #include <algorithm> std::vector<int> vec = { ... }; std::sort(vec.begin(), vec.end());
- 13. Sort Host Data #include <vector> #include <boost/compute/algorithm/sort.hpp> std::vector<int> vec = { ... }; boost::compute::sort(vec.begin(), vec.end(), queue); 0 2000 4000 6000 8000 1M 10M 100M STL Boost.Compute
- 14. Parallel Reduction #include <boost/compute/algorithm/reduce.hpp> #include <boost/compute/container/vector.hpp> boost::compute::vector<int> data = { ... }; int sum = 0; boost::compute::reduce( data.begin(), data.end(), &sum, queue ); std::cout << “sum = “ << sum << std::endl;
- 15. Algorithm Internals • Fundamentally, STL-like algorithms produce OpenCL kernel objects which are executed on a compute device. C++ OpenCL
- 16. Custom Functions BOOST_COMPUTE_FUNCTION(int, plus_two, (int x), { return x + 2; }); boost::compute::transform( v.begin(), v.end(), v.begin(), plus_two, queue );
- 17. Lambda Expressions using boost::compute::lambda::_1; boost::compute::transform( v.begin(), v.end(), v.begin(), _1 + 2, queue ); • Offers a concise syntax for specifying custom operations • Fully type-checked by the C++ compiler
- 19. OpenGL Interop • OpenCL provides mechanisms for synchronizing with OpenGL to implement direct rendering on the GPU • Boost.Compute provides easy to use functions for interacting with OpenGL in a portable manner. OpenGLOpenCL
- 20. Program Caching • Helps mitigate run-time kernel compilation costs • Frequently-used kernels are stored and retrieved from the global cache • Offline cache reduces this to one compilation per system
- 21. Auto-tuning • OpenCL supports a wide variety of hardware with diverse execution characteristics • Algorithms support different execution parameters such as work-group size, amount of work to execute serially • These parameters are tunable and their results are measurable • Boost.Compute includes benchmarks and tuning utilities to find the optimal parameters for a given device
- 22. Auto-tuning
- 23. Recent News
- 24. Coming soon to Boost • Went through Boost peer-review in December 2014 • Accepted as an official Boost library in January 2015 • Should be packaged in a Boost release this year (1.59)
- 25. Boost in GSoC • Boost is an accepted organization for the Google Summer of Code 2015 • Last year Boost.Compute mentored a student who implemented many new algorithms and features • Open to mentoring another student this year • See: https://svn.boost.org/trac/boost/wiki/SoC2015