Slow smartlog re-indexing

[cem82]

I'm needing to re-index a lot of log files however it is running very slowly (freshly copied from another server).  We're running R81.20 JHF 41 which is a fresh install and have 32 CPU cores / 64GB RAM (10G used).  However all CPU cores are running low utilisation (most are 95% idle and a handful around 30%) and no IO wait time.  It's taking around 40+ min per 2GB log file to index so wondering if there's anything we can do to speed that up?  Currently we have logging going to different MLM log servers while the reindexing is happening with the intention for it to complete as fast as possible.

[PhoneBoy]

By default, (re)indexing processes are started with lowest priority (39) by design.
This means any other use of CPU will be given priority.
You can use the standard "renice" command in Linux to increase the priority (20 is "standard" priority, or you can make it higher).