Archive Performance

Question

Hi there,
In a standby database configuration I'm using the following query to determine the MB per second of archivelogs being written to the first five archivelog destinations.
select next_time,
       nvl(avg(case when ad.dest_name = 'LOG_ARCHIVE_DEST_1' and (completion_time - next_time) * 24 * 60 * 60 > 0
            then (blocks * block_size/1024/1024) / ((completion_time - next_time) * 24 * 60 * 60) end),0) as DEST_1,
       nvl(avg(case when ad.dest_name = 'LOG_ARCHIVE_DEST_2' and (completion_time - next_time) * 24 * 60 * 60 > 0
            then (blocks * block_size/1024/1024) / ((completion_time - next_time) * 24 * 60 * 60) end),0) as DEST_2,
       nvl(avg(case when ad.dest_name = 'LOG_ARCHIVE_DEST_3' and (completion_time - next_time) * 24 * 60 * 60 > 0
            then (blocks * block_size/1024/1024) / ((completion_time - next_time) * 24 * 60 * 60) end),0) as DEST_3,
       nvl(avg(case when ad.dest_name = 'LOG_ARCHIVE_DEST_4' and (completion_time - next_time) * 24 * 60 * 60 > 0
            then (blocks * block_size/1024/1024) / ((completion_time - next_time) * 24 * 60 * 60) end),0) as DEST_4,
       nvl(avg(case when ad.dest_name = 'LOG_ARCHIVE_DEST_5' and (completion_time - next_time) * 24 * 60 * 60 > 0
            then (blocks * block_size/1024/1024) / ((completion_time - next_time) * 24 * 60 * 60) end),0) as DEST_5
from v$archived_log al,
     v$archive_dest ad
where al.dest_id = ad.dest_id
group by next_time
order by next_time;The logic behind the script depends on the fact that the next_time column in the v$archived_log table refers to the point in time when the redo log is switched and writing to the archive log destinations begin, can anyone confirm if this is the case? If not, what is the best way to determine the performance of writing the archive destinations?
Note: I know the script is not optimised, I want to get the logic correct before progressing to performance etc.
Thank you
ezuall

Answer 1

Larry,
Thank you for the feedback.
The first destination is local and the other 4 are remote standby destinations.
I am trying to work out at what rate (in MB per second) archivelogs are being written to each destination. I'm hoping this will help me to get an overview of how well the standby destinations are working and how much redo could be created in the Primary Database before the standby databases can no longer keep up.
Regards
ezuall