I ended with 125384 entries like the one above.

Everything works fine but ......... [U]tteerrrriibbllee[/U] [B]slow[/B].  CPU usage is constant at 99%.  

I tried to split the one file produced by the import into 2 chunks but that did not improve the situation.

My question is : Is there a rule of thumb for the size of the "database" in which elog keeps it's performance, or better : is there a solution for this for elog.