Introduction to Parallel Programming

Preview 11:31

Parallel Architectures and Programming Models

Preview 11:25

Pipelining

Preview 09:02

Superpipelining and VLIW

Preview 13:21

Memory Latency

Preview 08:37

Cache and Temporal Locality

Preview 14:43

Cache, Memory bandwidth and Spatial Locality

Preview 12:23

Intuition for Shared and Distributed Memory architectures

Preview 16:58

Shared and Distributed Memory architectures

Preview 04:24

Interconnection networks in Distributed Memory architectures

Preview 15:12

OpenMP: A parallel Hello World Program

Preview 05:15

Program with Single thread

Preview 13:56

Program Memory with Multiple threads and Multi-tasking

Preview 07:20

Context Switching

Preview 12:11

OpenMP: Basic thread functions

Preview 09:52

OpenMP: About OpenMP

Preview 05:13

Shared Memory Consistency Models and the Sequential Consistency Model

Preview 21:33

Race Conditions

Preview 06:14

OpenMP: Scoping variables and some race conditions

Preview 15:55

OpenMP: thread private variables and more constructs

Preview 11:14

Computing sum: first attempt at parallelization

Preview 08:38

Manual distribution of work and critical sections

Preview 10:52

Distributing for loops and reduction

Preview 06:39

Vector-Vector operations (Dot product)

Preview 19:36

Matrix-Vector operations (Matrix-Vector Multiply)

Preview 14:38

Matrix-Matrix operations (Matrix-Matrix Multiply)

Preview 20:30

Introduction to tasks

Preview 07:07

Task queues and task execution

Preview 10:53

Accessing variables in tasks

Preview 11:12

Completion of tasks and scoping variables in tasks

Preview 08:49

Recursive task spawning and pitfalls

Preview 11:09

Understanding LU Factorization

Preview 15:17

Parallel LU Factorization

Preview 14:12

Locks

Preview 23:41

Advanced Task handling

Preview 15:12

Matrix Multiplication using tasks

Preview 05:02

The OpenMP Shared Memory Consistency Model

Preview 14:59

Applications finite element method

Preview 32:55

Applications deep learning

Preview 26:47

Introduction to MPI and basic calls

Preview 09:19

MPI calls to send and receive data

Preview 07:30

MPI calls for broadcasting data

Preview 23:33

MPI non blocking calls

Preview 13:38

Application distributed histogram updation

Preview 23:34

MPI collectives and MPI broadcast

Preview 10:47

MPI gathering and scattering collectives

Preview 05:47

MPI reduction and alltoall collectives

Preview 08:41

Discussion on MPI collectives design

Preview 09:54

Characteriziation of interconnects

Preview 07:11

Linear arrays 2D mesh and torus

Preview 10:11

d dimensional torus

Preview 08:28

Hypercube

Preview 09:49

Trees and cliques

Preview 08:00

Hockney model

Preview 08:43

Broadcast and Reduce with recursive doubling

Preview 14:56

Scatter and Gather with recursive doubling

Preview 05:39

Reduce scatter and All gather with recursive doubling

Preview 13:46

Discussion of message sizes in analysis

Preview 03:30

Revisiting Reduce scatter on 2D mesh

Preview 06:20

Reduce scatter and Allreduce on the Hypercube

Preview 15:23

Alltoall on the Hypercube

Preview 08:45

Lower bounds

Preview 07:43

Pipeline based algorithm for Allreduce

Preview 11:18

An improved algorithm for Alltoall on the Hypercube using E cube routing

Preview 20:32

Pipeline based algorithm for Broadcast

Preview 04:28

Introduction to parallel graph algorithms

Preview 22:55

Breadth First Search BFS using matrix algebra

Preview 17:54

BFS Shared memory parallelization using OpenMP

Preview 24:18

Distributed memory settings and data distribution

Preview 09:45

Distributed BFS algorithm

Preview 21:35

Performance considerations

Preview 07:41

Prims Algorithm

Preview 12:28

OpenMP based shared memory parallelization for MST

Preview 19:24

MPI based distributed memory parallelization for MST

Preview 11:49

Sequential Algorithm Adaptation from Prims

Preview 10:28

Parallelization Strategy for Prims algorithm

Preview 26:03

Dry run with the parallel strategy

Preview 18:03

Johnsons algorithm with 1D data distribution

Preview 10:17

Speedup analysis on a grid graph

Preview 09:27

Floyds algorithm for all pair shortest paths

Preview 12:32

Floyds algorithm with 2D data distribution

Preview 21:02

Adaptation to transitive closures

Preview 07:54

Parallelization strategy for connected components

Preview 26:14

Analysis for parallel connected components

Preview 11:18