Normal view MARC view ISBD view

PyGraph : compiler support for efficient and transparent use of CUDA graphs

By:

Ghosh, Abhishek

Contributor(s):

Advised by Basu, Arkaprava

Material type: Text

TextLanguage: en Publication details: Bangalore : Indian Institute of Science. 2024.Description: vii, 41 p. : col. ill. e-Thesis 1.670 MbSubject(s):

DDC classification:

621.39 BAS

Online resources:

Click here to access online

Dissertation note: MTech(Res);2025;Computer Science and Automation. Summary: CUDA Graphs --- a recent hardware feature introduced for NVIDIA GPUs --- aim to reduce CPU launch overhead by capturing and launching a series of GPU tasks (kernels) as a DAG. However, deploying CUDA Graphs faces several challenges today due to the static structure of a graph. It also incurs performance overhead due to data copy. In fact, we show a counter-intuitive result --- deploying CUDA Graphs hurts performance in many cases.

Tags from this library: No tags from this library for this title. Log in to add tags.

Average rating: 0.0 (0 votes)

Holdings
Item type	Current library	Call number	URL	Status	Date due	Barcode
Thesis	JRD Tata Memorial Library	621.39 BAS (Browse shelf(Opens below))	Link to resource	Not for loan		ET00791

Includes bibliographical references.

MTech(Res);2025;Computer Science and Automation.

CUDA Graphs --- a recent hardware feature introduced for NVIDIA GPUs --- aim to reduce CPU launch overhead by capturing and launching a series of GPU tasks (kernels) as a DAG. However, deploying CUDA Graphs faces several challenges today due to the static structure of a graph. It also incurs performance overhead due to data copy. In fact, we show a counter-intuitive result --- deploying CUDA Graphs hurts performance in many cases.

There are no comments on this title.

to post a comment.