X hits on this document

Powerpoint document

The Missouri S&T CS GPU Cluster - page 8 / 18

62 views

0 shares

0 downloads

0 comments

8 / 18

The CUDA programming model

A grid is organized into blocks, and each block is organized into threads.

Only threads within the same block can communicate via shared memory; and sync.

This type of organization helps the GPU parallelize thread execution using it’s inbuilt hardware protocols.

Document info
Document views62
Page views62
Page last viewedMon Jan 23 21:06:52 UTC 2017
Pages18
Paragraphs213
Words1253

Comments