GPU Programming NSM Nodal Centre for Training in HPC and AI

Slides: Intro + Logistics Computation Memory Synchronization Functions Support Streams Topics Case Study -- Graphs

Resources: All codes Lecture videos FAQ one and two Compute capability crossword Doubts Session 1: February 8 Doubts Session 2: March 9 Doubts Session 3: April 1 Doubts Session 4: May 4 Assignment 1 on Computation: Statement, Header file, Sample testcases Assignment 2 on Memory: Statement, main.cu, Sample testcases Assignment 3 on Synchronization: Statement, main.cu, Sample testcases Assignment 4 on Application: Statement, Sample testcases

Recorded Lectures

Month	Dates	Topic	Comments
February	1, 2, 3, 5	Introduction, Computation	Hello World, One, Two, Three Grid, Blocks, Threads Kernel Launch: 1D, 1D-General, 2D
	8, 9, 10, 12	Computation	CPU-GPU Communication (cudaMalloc, cudaMemcpy) Global variables Matrix mult.: CPU, Outer parallel, Outer+Inner parallel
	15, 16, 17, 19	Computation	Thread Divergence Divergence due to switch Problem Set 1
	22, 23, 24, 26	Memory	Memory Coalescing AoS versus SoA Barrier
March	1, 2, 3, 5	Memory, Support	Linked List Copying Shared Memory Shared Memory with Barrier String Permutation Dynamic Shared Memory Dynamic Shared Memory with Multiple Arrays CUDA GDB Error Handling Dangling Pointer
	8, 9, 10, 12	Memory	Texture Memory (via CUDA SDK) Constant Memory Bank Conflicts Problem Set 2 NvProf Original Code Loop Fusion Kernel Fusion Converting Loop to Blocks
	15, 16, 17, 19	Synchronization	Convolution Worklist Insertion Task Donation
	22, 23, 24, 26	Synchronization	Reduction: i + N/2, N - i, i + 1 Prefix Sum / Scan
	29, 30, 31	Synchronization	No classes due to semester break. No Global Barrier Global Barrier using Atomics Hierarchical Global Barrier
April	2	Synchronization	No classes due to semester break. Linked List Insertion CPU-GPU Shared Pinned Memory Persistent Kernel Problem Set 3
	5, 6, 7, 9	Functions	Array increment: Sequential, Parallel Thrust basics Thrust Reduction Thrust Prefix Sum Thrust-like device vector implementation
	12, 13, 14, 16	Streams	Basic Stream Program with Asynchronous memcpy with cudaHostAlloc Cooperative Kernels
	19, 20, 21, 23	Topics	Dynamic Parallelism Conditional Child Kernels using Global Device Memory with Non-Blocking Streams
	26, 27, 28, 30	Topics	MultiGPU: Number of Devices Cross-Device Synchronization PTX: CUDA Code, Assembly Code Basic Warp Voting Converting Mask to Count (popc) Use of ffs Conditional Participation in ballot
May	3, 4, 5, 7	Topics, Case Study	Loop Unrolling, Unrolled Assembly Heterogeneous Computation with Shared Variable Task Distribution OpenMP Reduction with HostAlloc'ed Memory Dynamic Scheduling OpenCL: Driver, Kernel