Gurus, I am running 3 imports of tables which have 14 mln, 14 mln and 17 mln rows respectively. Tables have similar columns on them. Here is the parfile :- cat impt06.par userid=username/password file=expt06.dmp full=y ignore=y feedback=10000 resumable=y log=impt06.log commit=y indexes=n I have done the following so far 1) Turned the archive logging off . (Yes isn't it great I can do that ?) 2) 'Niced' the process priority up 3) Added commit=y 4) Indexes =n 5) Added a buffer of 92160 (default multiply by 3) Now does the resumable parameter do something that will add time to imports ? I am on 9.2.0.1 and am on HPUX 11.11 Here are a few of the parameters that I am thinking about tweaking. From db_cache_size=57671680 To db_cache_size=157286400 db_file_multiblock_read_count=16 -- Should I even look at this ? java_pool_size=115343360 large_pool_size=12582912 shared_pool_size=314572800 Please respond to either the list enmasse or to taptriv@xxxxxxxxx Sincerely, Tapan Trivedi Tapan Trivedi Senior Oracle DBA 916 613 1921 test'; "> test'; ">