Numerical methods and HPC
Open Access

Fig. 14

image

Download original image

Performance (speedup over sequential execution) achieved for computing from 256 to 256 000 neurons of size 512 using different morphologies on multicore architecture (Multicore in the graph) and GPU (Multi in the graph) respectively. We have also included the mono-morphology approach (Mono in the graph) for a batch of the same morphology of size 512.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.