No
Yes
View More
View Less
Working...
Close
OK
Cancel
Confirm
System Message
Delete
Schedule
An unknown error has occurred and your request could not be completed. Please contact support.
Scheduled
Wait Listed
Personal Calendar
Speaking
Conference Event
Meeting
Interest
There aren't any available sessions at this time.
Conflict Found
This session is already scheduled at another time. Would you like to...
Loading...
Please enter a maximum of {0} characters.
{0} remaining of {1} character maximum.
Please enter a maximum of {0} words.
{0} remaining of {1} word maximum.
must be 50 characters or less.
must be 40 characters or less.
Session Summary
We were unable to load the map image.
This has not yet been assigned to a map.
Search Catalog
Reply
Replies ()
Search
New Post
Microblog
Microblog Thread
Post Reply
Post
Your session timed out.
This web page is not optimized for viewing on a mobile device. Visit this site in a desktop browser to access the full set of features.
GTC 2018 Silicon Valley

S8109 - Kokkos' Task DAG Parallelism for C++ HPC Applications

Session Speakers
Session Description

Kokkos' directed acyclic graph of tasks (Task DAG) capability has finally arrived. Previously, Kokkos only provided HPC C++ applications with performance portable data parallel patterns (foreach, reduce, and scan). After several years of challenging R&D, we have deployed support for Task DAG patterns on NVIDIA GPU, Intel Xeon Phi, and multicore CPUs. Kokkos now provides applications with a straightforward and architecture-independent interface for implementing their Task DAG algorithms. We present the dynamic heterogeneous "Task-DAG" capability and its use in Tacho for a multifrontal sparse matrix factorization library. We also present the new static homogeneous "work-DAG" pattern and its use in Tycho 2 for a neutral particle transport via sweeps algorithm.


Additional Information
Tools and Libraries, Supercomputing
General
All technical
Talk
50 minutes
Session Schedule