I recently updated our production EPE HIBP DB on our new servers and noticed that it still took a long time to update the DB. I checked our processors during this and noticed that when it’s uploading the HIBP hashes to the Enterprise Manager from a local password.txt file, it seems to peg only one of my logical processors, and I have 64.
Is there a way to ensure that when installing or upgrading or anything of that nature, the process would allow us to spread the load across all of the processors to speed up the install or update?
Hi Jay - I need to confirm with dev but i believe the process of updating the 64k small hash files we create for faster lookup from the monolithic HIBP content is single threaded, thus can only use one core. It would be a significant re-write to make it multi-threaded at which point would likely become I/O bound and not really be any faster.
