Skip to main content
Fig. 7 | BMC Bioinformatics

Fig. 7

From: Locality-sensitive hashing enables efficient and scalable signal classification in high-throughput mass spectrometry raw data

Fig. 7

Scalability for different m and n: wall-clock time of the implementation in seconds as a function of the number of threads used. More precisely, the cumulative runtime for processing of 100 randomly selected precursor frames is shown. Note the logarithmic scales on both axes. First the wall-clock time is a decreasing function of the number of threads used and then saturates. This shows that the implementation scales out well up to a certain point, at which the parallelization overhead cancels the gain in speed. See the text for the setup used

Back to article page