Gurus, Thank you all for the insights.. I will proceed with these thoughts Thanks again Bala On Fri, Aug 2, 2013 at 3:43 PM, Tim Gorman <tim@xxxxxxxxx> wrote: > Either EXPDP from the source database, or IMPDP from the destination > database, using the NETWORK_LINK and PARALLEL parameters. Prevents > time, space, and I/Os from being wasted on writing to and reading from > intermediate ".dmp" files. > > On 8/2/2013 1:37 PM, Bala wrote: > > I have a requirement to copy 80 mllion rows table from PROD to DEV. > > This table is 28 GB in size, has 4 indexes built on its columns. > > > > 11gR2 (11.2.0.3), Linux Redhat 6 environment. > > > > What would be the most optimal way to do it -- use traditional export or > > datapump (expdp) ,any insight will be greatly appreciated. > > > > Thank you all for your time. > > > > -- > //www.freelists.org/webpage/oracle-l > > > -- Bala Rao -- //www.freelists.org/webpage/oracle-l