Empno, :new.Ename, :new.Deptno,
:new.Salary, :new.Birth_Date, :new.Soc_Sec_Num);
end;
It is theoretically possible to create a trigger to replicate all possible permutations of datamanipulation
actions on the local database, but this quickly becomes difficult to manage. For
a complex environment, you should consider the use of materialized views. For the limited
circumstances described earlier, triggers are a very easy solution to implement.
NOTE
If you use triggers for your data replication, the success of a transaction
in the master database is dependent on the success of the remote
transaction.
Managing Materialized Views
You can use materialized views to aggregate, pre-join, or replicate data. In an enterprise database
environment, data generally flows from an online transaction-processing database into a data
warehouse. Normally the data is prestaged, cleansed, or otherwise processed and then moved
into the data warehouse. From there, the data may be copied to other databases or data marts.
You can use materialized views to pre-compute and store aggregate information within a
database, to dynamically replicate data between distributed databases, and synchronize data
updates within replicated environments. In replication environments, materialized views enable
local access to data that would normally have to be accessed remotely. A materialized view may
be based on another materialized view.
In large databases, materialized views help to improve the performance of queries that involve
aggregates (including sum, count, average, variance, standard deviation, minimum, and maximum)
or table joins.
Pages:
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899