010 - C++ CUDA ver. 11.2 Memory Pool, cudaMallocAsync, cudaFreeAsync, cudaMemcpyAsync

Описание

1. CUDA stream, create_stream()
2. Pinned Host Memory, cudaHostAlloc(), cudaMallocHost()
3. cudaMemcpyAsync()

네이티브가 아니면 은근 쉽게 쓰기 힘든 표현 3개(further along, reiterate, stick to)
https://tinyurl.com/47pdpkwt

Unified Memory in CUDA 6
https://developer.nvidia.com/blog/unified-memory-in-cuda-6/

Parallel Forall Code Samples
https://github.com/NVIDIA-developer-blog/code-samples

1. CUDA C/C++ Streams and Concurrency
https://tinyurl.com/ycks9s5c

2. Enhancing Memory Allocation with New NVIDIA CUDA 11.2 Features
https://tinyurl.com/5dmmnfvr

3. Using the NVIDIA CUDA Stream-Ordered Memory Allocator, Part 1
https://tinyurl.com/4z7sahy3

4. Using the NVIDIA CUDA Stream-Ordered Memory Allocator, Part 2
https://tinyurl.com/3v6w9cp6

5. Getting Started with CUDA Graphs
https://tinyurl.com/5ykcbkfb

6.4. Stream Management - https://tinyurl.com/2p882sx4
6.5. Event Management - https://tinyurl.com/6rpj3hcf
6.9. Memory Management - https://tinyurl.com/2ewt3mew
6.11. Stream Ordered Memory Allocator - https://tinyurl.com/y7n7c2xy

From Previous Videos:

010 - C++ CUDA ver. 11.2 Memory Pool, cudaMallocAsync, cudaFreeAsync, cudaMemcpyAsync
Source: https://sourcecode.talkplayfun.com/advanced-cuda/010-MemoryPool.zip

009 - cudaMemcpyAsync, cudaMallocHost or cudaHostAlloc for page-locked or pinned host memory
Video: https://www.youtube.com/watch?v=R6X6WCml-6U&list=PLsIvhalfft1Enlg4L77-sdLc_PPh6jH-b&index=9
Source: https://sourcecode.talkplayfun.com/advanced-cuda/009-PageLockedMemory.zip

008 - Asynchronous Multiple CUDA Kernels, cudaStreamWaitEvent(), cudaStreamSynchronize()
Video: https://www.youtube.com/watch?v=83k75_VnMaE&list=PLsIvhalfft1Enlg4L77-sdLc_PPh6jH-b&index=8
Source: https://sourcecode.talkplayfun.com/advanced-cuda/008-MultipleKernels.zip

007 - Null Stream, Non-Null Stream, create_event(), create_stream(), Fail-Safe Programming
Video: https://www.youtube.com/watch?v=kKmHa3mTLNY&list=PLsIvhalfft1Enlg4L77-sdLc_PPh6jH-b&index=7
Source: https://sourcecode.talkplayfun.com/advanced-cuda/007-NonNullStream.zip

006 - Template Template Parameter, for_array, for_tuple, cudaStream_t, cudaEvent_t, create_stream()
Video: https://www.youtube.com/watch?v=OcZZOb6lNq8&list=PLsIvhalfft1Enlg4L77-sdLc_PPh6jH-b&index=6
Source: https://sourcecode.talkplayfun.com/advanced-cuda/006-for_array.zip

005 - RAII for C++ CUDA, std::unique_ptr, cudaStream_t 완전 기초, create_event()
Video: https://www.youtube.com/watch?v=UItrKQ7nZFk&list=PLsIvhalfft1Enlg4L77-sdLc_PPh6jH-b&index=5
Source: https://sourcecode.talkplayfun.com/advanced-cuda/005-unique_ptr.zip

004 - C++ attribute: nodiscard 사용법, Warning As Error -Xcompiler=/we4834, RAII for C++ CUDA
Video: https://www.youtube.com/watch?v=4dEXU6PBTOU&list=PLsIvhalfft1Enlg4L77-sdLc_PPh6jH-b&index=4
Source: https://sourcecode.talkplayfun.com/advanced-cuda/004-nodiscard.zip