Skip to Main Content
Designing Scientific Applications on GPUs
book

Designing Scientific Applications on GPUs

by Raphael Couturier
November 2013
Intermediate to advanced content levelIntermediate to advanced
498 pages
17h 6m
English
Chapman and Hall/CRC
Content preview from Designing Scientific Applications on GPUs
Software components for heterogeneous many-core architectures 93
algorithms, that are well suited for such systems, forcing developers to search
for novel methods that utilize concurrency.
To ease software development, we use MPI-2 for message passing and en-
sure a safe and private communication space by creation of a communicator
private to the library during initialization, as recommended by Hoefler and
Snir [12]. With the addition of remote direct memory access (RDMA) for
GPUDirect it is possible to make direct memory transfers between recent gen-
eration of GPUs (Kepler), eliminating CPU overhead. Unfortunately there are
some strict system and driver requirements to enable these features. Therefore,
in the following examples, device memory is first ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Start your free trial

You might also like

Introduction to Numerical Analysis and Scientific Computing

Introduction to Numerical Analysis and Scientific Computing

Nabil Nassif, Dolly Khuwayri Fayyad
Computational Electromagnetism

Computational Electromagnetism

Alain Bossavit, Isaak D. Mayergoyz

Publisher Resources

ISBN: 9781466571648