You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We want to partition already at the build-hit stage (since hit parameters like energy resolution vary across partitions. Figuring out which file to open and from which row (step index) to start reading is challenging. We should think about an efficient way to compute a lookup table once per workflow and re-use it, considering that:
Geant4 event indices are not linearly increasing in output files with multithreaded remage
If we want to merge files from multiple threads (desired because of filesystem reasons), the indices won't be sorted either (unless we do some explicit sorting)
The output hit files will need to have an event ID structure that makes it easy to build events later
Let's think about a good solution here
The text was updated successfully, but these errors were encountered:
We want to partition already at the build-hit stage (since hit parameters like energy resolution vary across partitions. Figuring out which file to open and from which row (step index) to start reading is challenging. We should think about an efficient way to compute a lookup table once per workflow and re-use it, considering that:
Let's think about a good solution here
The text was updated successfully, but these errors were encountered: