SCHEDULE: NOV 10-16, 2012
When viewing the Technical Program schedule, on the far righthand side is a column labeled "PLANNER." Use this planner to build your own schedule. Once you select an event and want to add it to your personal schedule, just click on the calendar icon of your choice (outlook calendar, ical calendar or google calendar) and that event will be stored there. As you select events in this manner, you will have your own schedule to guide you through the week.
Dataflow-Driven GPU Performance Projection for Multi-Kernel Transformations
SESSION: Performance Modeling
EVENT TYPE: Papers
TIME: 10:30AM - 11:00AM
SESSION CHAIR: Dimitris Nikolopoulos
AUTHOR(S):Jiayuan Meng, Vitali Morozov, Venkatram Vishwanath, Kalyan Kumaran
Applications often have a sequence of parallel operations to be offloaded to graphics processors; each operation can become an individual GPU kernel. Developers typically explore different transformations for each kernel. It is wellknown that efficient data management is critical in achieving high GPU performance and that “fusing” multiple kernels into one may greatly improve data locality. Doing so, however, requires transformations across multiple, potentially nested, parallel loops; at the same time, the original code semantics must be preserved. Since each kernel may have distinct data access patterns, their combined dataflow can be nontrivial. As a result, the complexity of multi-kernel transformations often leads to significant effort with no guarantee of performance benefits. This paper proposes a dataflow-driven analytical framework to project GPU performance for a sequence of parallel operations without implementing GPU code or using physical hardware. The framework also suggests multi-kernel transformations that can achieve the projected performance.
Dimitris Nikolopoulos (Chair) - Queen's University, Belfast
Jiayuan Meng - Argonne National Laboratory
Vitali Morozov - Argonne National Laboratory
Venkatram Vishwanath - Argonne National Laboratory
Kalyan Kumaran - Argonne National Laboratory