SCHEDULE: NOV 16-22, 2013
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.
Algorithmic Choice in Optimization Problems: A Performance Study
SESSION: Research Poster Reception
EVENT TYPE: Posters, Electronic Posters, and Education Posters
TIME: 5:15PM - 7:00PM
AUTHOR(S):Hammad Rashid, Clara Novoa, Richard Hay, Apan Qasem
ROOM:Mile High Pre-Function
ABSTRACT:
To harness the full potential of emerging manycore platforms, the compiler needs to play a key role in exploiting the available on-chip parallelism. Since the amount of extracted parallelism is directly influenced by the selection of the algorithm, algorithmic choice plays a critical role in achieving scalable high performance. This research investigates the impact of algorithmic choice on
performance of parallel implementations of optimization problems. The study implements several algorithmic variants of the integral knapsack problem and evaluates each one based on a range of performance characteristics including thread affinity, task granularity and data reuse. Experimental results reveal that selection of the algorithm does have a significant impact on parallel performance. Furthermore, the study provides insight into the relationship of algorithm selection and specific aspects of performance including HW prefetch activity and exploited data locality through favorable cache sharing.
Chair/Author Details:
Hammad Rashid - Texas State University
Clara Novoa - Texas State University
Richard Hay - Texas State University
Apan Qasem - Texas State University
Click here to download .ics calendar file
