SC10 Coprocessing Tutorial: Difference between revisions
No edit summary |
No edit summary |
||
| Line 1: | Line 1: | ||
''In-situ'' visualization is a term for running a solver in tandem with visualization. By coupling these together we can utilize the high performance computing for post processing, and we can circumvent the bottlenecks associated with storing and retrieving data in disk storage. To simplify the integration of visualization and post processing into computational code, we present the coprocessing library provided as part of the ParaView framework. ParaView is a powerful open-source turnkey application for analyzing and visualizing large data sets in parallel. The coprocessing library provides a programmatic interface to the ParaView framework that simplifies integration with existing codes. Attendees will learn the structure of the coprocessing API and how to bind it to C, C++, | ''In-situ'' visualization is a term for running a solver in tandem with visualization. By coupling these together we can utilize the high performance computing for post processing, and we can circumvent the bottlenecks associated with storing and retrieving data in disk storage. To simplify the integration of visualization and post processing into computational code, we present the coprocessing library provided as part of the ParaView framework. ParaView is a powerful open-source turnkey application for analyzing and visualizing large data sets in parallel. The coprocessing library provides a programmatic interface to the ParaView framework that simplifies integration with existing codes. Attendees will learn the structure of the coprocessing API and how to bind it to C, C++, Fortran, and Python. Attendees will also receive instructions on customizing the ParaView build to a particular code, minimizing the memory footprint, and simplifying the specification of analysis tasks via integration with the ParaView application. | ||
This tutorial is being held on the afternoon of Sunday, November 14 as part of the [http://sc10.supercomputing.org/ Supercomputing 2010 Conference]. See the [http://sc10.supercomputing.org/schedule/event_detail.php?evid=tut110 Supercomputing tutorial page] for more information on the time, the location, and to register (be sure to register for the Sunday tutorials). | This tutorial is being held on the afternoon of Sunday, November 14 as part of the [http://sc10.supercomputing.org/ Supercomputing 2010 Conference]. See the [http://sc10.supercomputing.org/schedule/event_detail.php?evid=tut110 Supercomputing tutorial page] for more information on the time, the location, and to register (be sure to register for the Sunday tutorials). | ||
*'''[[Media:SC10 tutorial full.pptx|Slides presented at the tutorial]]''' | |||
*'''[[Media:SC10 tutorial examples.tar.gz|Code examples]]''' | |||
Latest revision as of 19:24, 17 September 2010
In-situ visualization is a term for running a solver in tandem with visualization. By coupling these together we can utilize the high performance computing for post processing, and we can circumvent the bottlenecks associated with storing and retrieving data in disk storage. To simplify the integration of visualization and post processing into computational code, we present the coprocessing library provided as part of the ParaView framework. ParaView is a powerful open-source turnkey application for analyzing and visualizing large data sets in parallel. The coprocessing library provides a programmatic interface to the ParaView framework that simplifies integration with existing codes. Attendees will learn the structure of the coprocessing API and how to bind it to C, C++, Fortran, and Python. Attendees will also receive instructions on customizing the ParaView build to a particular code, minimizing the memory footprint, and simplifying the specification of analysis tasks via integration with the ParaView application.
This tutorial is being held on the afternoon of Sunday, November 14 as part of the Supercomputing 2010 Conference. See the Supercomputing tutorial page for more information on the time, the location, and to register (be sure to register for the Sunday tutorials).
