|
MGA Duplica White Paper - Version 3.01
View
in Microsoft Word format (1,037 kb)
View
in PDF format (546 kb)
Download
in Microsoft Word format (contained in Zip file 224 kb)
Download
in PDF format (contained in Zip file 292 kb)
MGA Duplica F.A.Q.
If you cannot find answers here please go to the MGA
Duplica support page.
Or if you would like to make a comment on our product,
or discuss your issues with other users of MGA Duplica, please post your
message on our MGA Forums.
Questions:
Can
Duplica replicate data from multiple source database?
Yes - User can pick and choose the database (by specifying
the database link name) to be the source for each group of tables. Duplica
can also replicate data to tables that belong to different users.

Yes - Duplica uses the Oracle supplied package, DBMS_JOB,
to schedule its background jobs. Jobs are submitted for the control mechanism
and for the suites (group of tables) that are scheduled to run. Once submitted,
these jobs will be managed by Oracle RDBMS engine, which executes them
at the specified interval in parallel sessions that are independent from
each other.

Yes - Duplica allows the user to specify additional
query predicate (in a form of a WHERE condition) to eliminate unwanted
rows, thus further qualifying replicated data. Duplica replicates all
columns that exist in the target table. Columns that exist in the source
table but not in the target table will be ignored

Yes - Duplica performs satisfactorily at all of our
clients' sites. Its performance varies vastly from site to site and from
application to application due to a number of factors. Some of these are:
- Physical distance between the databases;
- The speed and the traffic of the communication link;
- The rate of changes at different times of the day when replication
occurs;
- The database configuration;
- The size of the table being replicated; and
- The replication frequency.
The average timings are roughly between 10,000 records
per minute to 100,000 records per minute on tables with approximately
3 million rows.
<< back to top
Yes - How long it takes for data to arrive at the target
database largely depends on the frequency of replication, which could
vary from seconds to days. Acceptable delay can also vary from table to
table and needs to be determined by the user based on the intended use
of the replicated changes. Replication frequency will then be set accordingly
via the GUI tool.
|