The "AWR" (Automatic Workload Repository) remains the cornerstone of this philosophy. However, in 12.2, the granularity of data has improved. DBAs should utilize the feature to consolidate performance data from multiple Pluggable Databases (PDBs) into a single repository. This allows for comparative analysis across your container database (CDB) environment, ensuring that "noisy neighbor" issues—where one PDB consumes resources and degrades the performance of others—are easily identified.
(The argument is seconds to analyze; 240 = last 4 minutes).
ALTER SYSTEM SET DB_PERFORMANCE_PROFILE = 'LOW' SCOPE = BOTH;
SQL tuning remains the bread and butter of performance optimization. In 12c Release 2, the optimizer has become significantly smarter, but it still requires human intervention for outlier cases. This allows for comparative analysis across your container
Most "tuning tips" recycle 10g indexing. 12.2 changes the math.
EXEC DBMS_STATS.SET_GLOBAL_PREFS('AUTO_STATS_GATHERING', 'ON');
Efficient SQL execution is the key to minimizing response times. In 12c Release 2, the optimizer has become
Automatically moves data between storage tiers based on access patterns, ensuring "hot" data stays on the fastest disks.
Before tuning, understand the engine. 12.2 introduced the as the default. If you are still tuning CDB$ROOT and PDBs like separate Oracle 11g instances, you are wasting resources.
: Configuring the database to automatically implement tuning tasks and verify results. System Scalability System Scalability In 12.1
In 12.1, SQL Plan Directives were often the root cause of sudden plan changes. 12.2 introduced automatic maintenance to purge stale directives. However, you should manually review:
However, simply enabling it isn't enough. In 12.2, utilize . For data that is read-only or infrequently updated, use higher compression levels (like MEMCOMPRESS FOR QUERY HIGH ) to fit more data into the IMCS, thereby reducing the bottleneck on the buffer cache.