Define 'long time' ... 15 mins? 2 hrs? 2 days?
70 million (crore = 10m ?) rows shouldn't take too long to copy unless you're experiencing some really lousy disk performance issues or you've run out of space.
A few things I'd want to check:
------------------------------
select * from master..monProcessStatement where SPID = <spid_of_reorg_command>
select * from master..monProcessActivity where SPID = <spid_of_reorg_command>
select * from master..monProcessWaits where SPID = <spid_of_reorg_command>
- run a few times at 1min intervals
- are any of the numbers (logical IOs, physical IOs, cpu counter) moving and if so, what kind of rates (eg, if physicals IOs are the only noticeable changes and they're relatively small => may indicate a performance issue with disk performance]
- if most of the interval time is taken up by waits, which waits event id's are showing the largest increases?
------------------------------
select * from master..monProcessStatement [*no* WHERE clause]
- how many other processes are running at the same time? [wondering if there could be contention for dataserver resources by 'too many' active processes]
------------------------------
sp_sysmon "00:10:00",noclear
- run while the reorg is running
- looking for indications of performance degradation, over-subscription to some dataserver resources, etc.
NOTE: If you decide to post the output from sp_sysmon to this thread, please place in a *txt file and attach said file to the thread (it's much easier to read from a *txt file as opposed to a really, Really, REALLY long cut-n-paste post in this thread!)
------------------------------
sp_who
- is the reorg command showing up in a LOG SUSPEND state (ie, log is full)?
- is the reorg command showing up as blocked?
------------------------------
check the ASE errorlog for any errors/messages that might indicate any issues
------------------------------
sp_monitorconfig 'all'
- any measurements showing up as 100% utilization and non-zero re-use counts?