Memory Structure distribution principle of dedicated Oracle servers

Generally, we will use the memory of the oracle instance to 80%-60% of the system's physical memory (4 GB memory or above used more than 80%, 2 GB memory used 60%-70%) spga + sga_max_size80 % mem

Generally, we will use the memory of the oracle instance to 80%-60% of the system's physical memory (4 GB memory or above used more than 80%, 2 GB memory used 60%-70%) s = pga + sga_max_size = 80% mem

Generally, we will use the memory of the Oracle instance to 80%-60% of the physical memory of the system (4G memory or above uses more than 80%, 2g memory uses 60%-70%)
S = pga + sga_max_size = 80% memory)
Db_cache_size = 60% sga
Shared_pool_size = sga_max_size-db_cache_size-160m can also be set to around 35% sga
Java_pool_size = 64 m
Large_pool_size = 16 m
Log_buffer = 2 m

1: 4 GB physical memory

If it is 4000 m
Then
S = 3200 m
For OLTP (
Sga = s * 80% = 2560 m
Db_cache_size = 1536 m data access buffer, dirty data before data submission is updated, etc.
Shared_pool_size = 864 m place SQL, plsql code (to improve code sharing) and data dictionary information (to Improve the access speed of data dictionary Information)
Java_pool_size = 64 m java application
Large_pool_size = 16 m used for backup and recovery, sharing the user session information (UGA) on the server (storing the user session information in pga on the dedicated server)
Log_buffer = 2 MB for log data buffering and Archiving
PGA = 640 M)

For DSS (
Sga = s * 50% = 1600 m
Db_cache_size = 960 m Data Access buffer, dirty data before data submission is updated, etc.
Shared_pool_size = 480 m place SQL, plsql code (to improve code sharing) and data dictionary information (to Improve the access speed of data dictionary Information)
Java_pool_size = 64 m java application
Large_pool_size = 16 m used for backup and recovery, sharing the user session information (UGA) on the server (storing the user session information in pga on the dedicated server)
Log_buffer = 2 MB for log data buffering and Archiving
PGA = 800 M)

2: 2 GB physical memory

If it is 2000 m
Then
S = 1200 m
For OLTP (
Sga = s * 80% = 960 m
Db_cache_size = 576 m data access buffer, dirty data before data submission is updated, etc.
Shared_pool_size = 224 m place SQL, plsql code (to improve code sharing) and data dictionary information (to Improve the access speed of data dictionary Information)
Java_pool_size = 64 m java application
Large_pool_size = 16 m used for backup and recovery, sharing the user session information (UGA) on the server (storing the user session information in pga on the dedicated server)
Log_buffer = 2 MB for log data buffering and Archiving
PGA = 240 M)

For DSS (
Sga = s * 50% = 600 m
Db_cache_size = 360 m Data Access buffer, dirty data before data submission is updated, etc.
Shared_pool_size = 144 m place SQL, plsql code (to improve code sharing) and data dictionary information (to Improve the access speed of data dictionary Information)
Java_pool_size = 32 m java application
Large_pool_size = 16 m used for backup and recovery, sharing the user session information (UGA) on the server (storing the user session information in pga on the dedicated server)
Log_buffer = 2 MB for log data buffering and Archiving
PGA = 240 M)

3: 8 GB physical memory

If it is 8000 m
Then
S = 6400 m
For OLTP (
Sga = s * 80% = 5120 m
Db_cache_size = 3072 m data access buffer, dirty data before data submission is updated, etc.
Shared_pool_size = 1864 m place SQL, plsql code (to improve code sharing) and data dictionary information (to Improve the access speed of data dictionary Information)
Java_pool_size = 64 m java application
Large_pool_size = 16 m used for backup and recovery, sharing the user session information (UGA) on the server (storing the user session information in pga on the dedicated server)
Log_buffer = 2 MB for log data buffering and Archiving
PGA = 1024 M)

For DSS (
Sga = s * 50% = 3200 m
Db_cache_size = 1920 m Data Access buffer, dirty data before data submission is updated, etc.
Shared_pool_size = 1120 m place SQL, plsql code (to improve code sharing) and data dictionary information (to Improve the access speed of data dictionary Information)
Java_pool_size = 64 m java application
Large_pool_size = 16 m used for backup and recovery, sharing the user session information (UGA) on the server (storing the user session information in pga on the dedicated server)
Log_buffer = 2 MB for log data buffering and Archiving
PGA = 3200 M)