UNDO Management and Dataguard questions

Hi guys,
I have two questions

1) Undo management = Manual / Auto.

When most documentations and Gurus (eg :TOM of Ask Tom ) Suggest usage of Undo tablespace to Auto. However in a huge transaction environment is it safe to use it ?
I remember vaguely reading up somewhere that when its set to automatic and when faced with huge transactions the extents sizes are allocated increased considerably i.e PCT_INCREASE 50% or so ?

Has anyone encountered such a situation before ? There were suggestions that the Undo be managed manaully .

2) In oracle 9i release 2.0 databases , what is the difference between standby ( physical ) and dataguard implementation ?

I had set the physical standby just as we used to do in the olden days when the new extra parametrs were not availble like fal_server and fal_Client . The logs are shipped and applied even now. However I could not set delay in the log_archive_dest_2 parameter... ?

So if i have the above setup how would i convert it to Dataguard ?? just setting up remaining parameter ? would it then be possible to do a easy switch over ?

Cheers
Dosth

1. Set it to AUTO. Monitor it. You'll be fine.
2. DG - you can set it up manually via command line DGMGRL. Or use OEM. OEM makes it terribly easy. (Although it is good idea to know how to do things via command line).