How branch mispredictions affect quicksort

Authors:
Kanela Kaligosi;Peter Sanders
Affiliations:
Max Planck Institut für Informatik, Saarbrücken, Germany;Universität Karlsruhe, Germany
Venue:
ESA'06 Proceedings of the 14th conference on Annual European Symposium - Volume 14
Year:
2006

Citing 4
Cited 3

Algorithm 64: Quicksort

Communications of the ACM
Optimal Sampling Strategies in Quicksort and Quickselect

SIAM Journal on Computing
Computer Architecture: A Quantitative Approach

Computer Architecture: A Quantitative Approach
Tradeoffs between branch mispredictions and comparisons for sorting algorithms

WADS'05 Proceedings of the 9th international conference on Algorithms and Data Structures

An experimental study of sorting and branch prediction

Journal of Experimental Algorithmics (JEA)
Algorithm engineering: bridging the gap between algorithm theory and practice

Algorithm engineering: bridging the gap between algorithm theory and practice
Branch mispredictions don't affect mergesort

SEA'12 Proceedings of the 11th international conference on Experimental Algorithms

Quantified Score

Hi-index	0.00

Visualization

Abstract

We explain the counterintuitive observation that finding "good" pivots (close to the median of the array to be partitioned) may not improve performance of quicksort. Indeed, an intentionally skewed pivot improves performance. The reason is that while the instruction count decreases with the quality of the pivot, the likelihood that the direction of a branch is mispredicted also goes up. We analyze the effect of simple branch prediction schemes and measure the effects on real hardware.