Oracle RAC allows multiple computers to run the Oracle RDBMS software simultaneously while accessing a single database, thus providing a clustered database environment. While there are many advantages of this system the main drawback is the complexity of the system to implement and administer. As this is such a specialised system we don’t have one at Borland to test upon. So the following suggestions are based on what we know of the system architecture. The architecture involves many machines each serving the purpose of either an Oracle Node or a Shared Storage server. If the purpose of monitoring is to test the throughput metrics for the database itself then it may be an option to monitor only the database and database machine using the built in Oracle support. Performance Explorer has built in support for Oracle 7.3, 8 and 9 although other versions can be monitored using V$SYSSTAT or SNMP. If the purpose is to monitor the whole RAC system including all nodes then we would suggest monitoring all other machines for Operating System Metrics only. This will at least provide a framework for which to improve upon in future monitoring. Options for monitoring unsupported versions of Oracle are SNMP, if an SNMP agent has been provided by Oracle, and V$SYSSTAT. For further information on these please see the shipped Performance Explorer Guide.
↧