Fig. 4: Runtime analysis. | Communications Chemistry

Fig. 4: Runtime analysis.

From: Ultra-large library screening with an evolutionary algorithm in Rosetta (REvoLd)

Fig. 4

The average runtime of REvoLd on different numbers of CPU cores on the high-performance compute cluster. Over 99% of all computational time is spent on docking (gray). More than half of the remaining time is spent loading and preparing the Enamine REAL space data (checkered). The evolutionary optimization (horizontal stripes) and setup of Rosetta (vertical stripes) is comparably fast and therefore neglectable. The setup time is barely visible within the figure and takes ~20 s, independent of the number of cores.

Back to article page