13: Parallele Algorithmen, Vorlesung, WS 2017/18, 29.01.2018 - a podcast by Karlsruher Institut für Technologie (KIT)

13 |
0:00:00 Starten
0:00:36 Was wissen wir über die Jobs?
0:02:32 Was wissen wir über die Prozessoren?
0:05:44 Zufälliges Zuordnen
0:07:08 Work Stealing
0:10:58 Backtracking over Transition Functions
0:12:02 An Abstract Model: Tree Shaped Computations
0:17:37 Splitting Stacks
0:21:27 Other Problem Categories
0:27:01 Limits of the Model
0:29:35 Receiver Initiated Load Balancing
0:31:40 Random Polling
0:41:11 Synchronous Random Polling
0:45:21 Analysis
0:51:22 Bounding Idleness
0:57:08 A Simplified Algorithm
1:03:22 Many Consecutive Splits
1:05:49 Many Processors
1:09:03 Superliner Speedup
1:15:12 Static vs Dynamic LB
1:18:35 MapReduce in 10 Minutes

Further episodes of Parallele Algorithmen, Vorlesung, WS17/18

Further podcasts by Karlsruher Institut für Technologie (KIT)

Website of Karlsruher Institut für Technologie (KIT)