Use the information in this appendix to select database templates for your planned databases.
Topics:
Choose a database template that best meets your database workload and hardware.
About Database Templates for Oracle Database Appliance
Oracle Database Appliance enables you to consolidate many databases into a single system. Consolidation can minimize idle resources, maximize efficiency, and lower costs. By using instance caging in conjunction with Oracle Database Resource Manager (the Resource Manager), you can provide desired levels of service across multiple instances on a single Oracle Database Appliance.
Oracle Database Appliance templates are already tuned for the size of each database instance workload. They are designed to run on a specific number of cores. Caging ensures that each database workload is restricted to the set of cores allocated by the template, enabling multiple databases to run concurrently with no performance degradation, up to the capacity of Oracle Database Appliance. You can select database template sizes larger than your current needs to provide for planned growth, which you accommodate later by adjusting System Global Area (SGA) and Program Global Area (PGA) sizes as well as the number of cores.
The Oracle Appliance Manager Configurator refers to the database sizing templates as classes of databases.
Note:
Oracle strongly recommends that you use the Oracle Database Appliance templates, because they implement best practices and are configured specifically for Oracle Database Appliance.
Use the following tables to help select the best templates for your databases. When using these tables remember that:
The information in the tables assumes that you are creating disk backups. The information in the tables assume that you are creating local disk backups. Consider the space requirements for your database and the policy for local disk backups versus external backups. Typically, external backups have more space available for the database than local backups.
Container databases are created on Oracle ACFS.
I/O per second (IOps) values are derived from an 8-kilobyte random read-write response time of 10 to 12 milliseconds for Hard Disk Drives (HDDs), and less than 1 millisecond for Flash, for Oracle Database Appliance X5-2; 5 to 7 milliseconds for Oracle Database Appliance X4-2 and X3-2; and 5 milliseconds for Oracle Database Appliance Version 1 to service an online transaction processing (OLTP) I/O request. The rates are not based on the number of CPUs and assume that the system is running at capacity.
Throughput, in megabytes per second (MBps) is based on a 1 MB sequential read/write for a data warehousing system. As with IOps, the MBps is a measure of throughput when the system is at capacity. With just a single small database running, the MBps would be the maximum available on the system for a large database.
The log file size assumes four redo log groups for each instance with a log switch every 15 minutes when the system is running at full capacity.
Storage is shared between the servers on Oracle Database Appliance.
About Improved Oracle Database Appliance Template Options
Oracle Appliance Manager release 12.1.2.2.0 and later releases provide improved database templates that are configured specifically for the type of database workload that you want to carry out on your databases on Oracle Database Appliance. Choose the template that best matches the common workload your databases perform (OLTP, DSS, In-Memory).
The improved Oracle Database templates replace the generic database templates from previous releases that only provided sizing parameters for OLTP workloads. The database sizing tables provide updated template names and sizing based on the number of CPUs and memory attributes for each type of database workload. The table “Oracle Database Appliance Database Template Name Conversions” provides a reference between the generic database template names and the database template names based on CPU cores.
Identify the template type that is appropriate to your database workload and hardware:
Use Oracle Database Appliance OLTP Database Templates if your database workload is primarily online transaction processing (OLTP).
Use Oracle Database Appliance DSS database templates if your database workload is primarily decision support services (DSS) or data warehousing.
Use Oracle Database Appliance In-Memory (IMDB) database templates if your database workload can fit in memory, and can benefit from in-memory performance capabilities.
Use Oracle Database Appliance X5-2 generic database templates if your database is on Oracle Database Appliance X5-2.
Use Oracle Database Appliance X4-2 generic database templates if your database is on Oracle Database Appliance X4-2.
Use Oracle Database Appliance X3-2 generic database templates if your database is on Oracle Database Appliance X3-2.
Use Oracle Database Appliance Version 1 generic database templates if your appliance is the original model of Oracle Database Appliance.
The templates provide general guidelines about the maximum number of databases on a fully deployed Oracle Database Appliance. Base the memory and CPU sizing on the number of instances per server, not the number of databases. Memory and CPUs are shared by all databases and the number of databases impacts performance. For example, when you have one database, the database receives full I/O throughput. If you have 9 databases, then each database has only 1/9th of the available I/O throughput for all disks.
When you have multiple databases, they share and compete over available CPU resources. When choosing a template, do not have more CPU_COUNTs distributed than available threads in the system. Oracle recommends partitioning, where every database has exclusive CPUs. If you choose to exceed the recommended maximum number of databases on Oracle Database Appliance (over-provision), be aware of your I/O results and performance degradation. With the over-provisioned approach, the sum of the CPU_COUNT across all instances can exceed the number of CPUs. Better resource utilization is provided with over-provisioning; however, it is possible for contention to occur when multiple databases are heavily loaded at the same time. Over-provisioning is best used for systems running non-critical applications. Avoid using over-provisioning for databases with high I/O requirements or those with high transaction rates. If you choose to over-provision, then do not exceed twice the number of total CPUs. Twice the allocation of total CPUs is based on hyper-threading of 2 CPU threads per core.
Table B-1 Oracle Database Appliance Database Generic Template Name Conversions
| CPU Core-Based Database Template Names | odb-01s | odb-01 | odb-02 | odb-04 | odb-06 | odb-12 | odb-16 | odb-24 |
|---|---|---|---|---|---|---|---|---|
|
Generic Database Template Names |
Very, very small |
Very Small |
Small |
Medium |
Large |
Extra Large |
Extra Extra Large |
Extra Extra Extra Large |
Note:
Extra Extra Large is applicable to Oracle Database Appliance X3-2 and Oracle Database Appliance X4-2. Extra Extra Extra Large is applicable to Oracle Database Appliance X4-2 only.
Table B-2 Oracle Database Appliance OLTP Database Template Sizes
| Template | CPU Cores | SGA | PGA | Flash | Processes | LOG buffer, Redo Log |
|---|---|---|---|---|---|---|
|
odb-01s (X5-2, X4-2, X3-2, and V1) |
1 |
2 GB |
1 GB |
6 GB |
200 |
16 MB, 1 GB |
|
odb-01 (X5-2, X4-2, X3-2, and V1) |
1 |
4 GB |
2 GB |
12 GB |
200 |
16 MB, 1 GB |
|
odb-02 (X5-2, X4-2, X3-2, and V1) |
2 |
8 GB |
4 GB |
24 GB |
400 |
16 MB, 1 GB |
|
odb-04 (X5-2, X4-2, X3-2, and V1) |
4 |
16 GB |
8 GB |
48 GB |
800 |
32 MB, 1 GB |
|
odb-06 (X5-2, X4-2, X3-2, and V1) |
6 |
24 GB |
12 GB |
72 GB |
1200 |
64 MB, 2 GB |
|
odb-12 (X5-2, X4-2, X3-2, and V1) |
12 |
48 GB |
24 GB |
144 GB |
2400 |
64 MB, 4 GB |
|
odb-16 (X5-2, X4-2, X3-2 Only) |
16 |
64 GB |
32 GB |
192 GB |
3200 |
64 MB, 4 GB |
|
odb-24 (X5-2, X4-2 Only) |
24 |
96 GB |
48 GB |
192 GB |
4800 |
64 MB, 4 GB |
|
odb-32 (X5-2 Only) |
32 |
128 GB |
64 GB |
256 GB |
6400 |
64 MB, 4 GB |
|
odb-36 (X5-2 Only) |
36 |
128 GB |
64 GB |
256 GB |
7200 |
64 MB, 4 GB |
Note:
Flash is applicable to Oracle Database Appliance X5-2 only.
Table B-3 Oracle Database Appliance DSS Database Template Sizes
| Template | CPU Cores | SGA (GB) | PGA (GB) | Processes | Redo log file size (GB) | Log buffer (MB) |
|---|---|---|---|---|---|---|
|
odb-01s (X5-2, X4-2, X3-2, and V1) |
1 |
1 |
2 |
200 |
1 |
16 |
|
odb-01 (X5-2, X4-2, X3-2, and V1) |
1 |
2 |
4 |
200 |
1 |
16 |
|
odb-02 (X5-2, X4-2, X3-2, and V1) |
2 |
4 |
8 |
400 |
1 |
16 |
|
odb-04 (X5-2, X4-2, X3-2, and V1) |
4 |
8 |
16 |
800 |
1 |
32 |
|
odb-06 (X5-2, X4-2, X3-2, and V1) |
6 |
12 |
24 |
1200 |
2 |
64 |
|
odb-12 (X5-2, X4-2, X3-2, and V1) |
12 |
24 |
48 |
2400 |
4 |
64 |
|
odb-16 (X5-2, X4-2, X3-2 Only) |
16 |
32 |
64 |
3200 |
4 |
64 |
|
odb-24 (X5-2, X4-2 Only) |
24 |
48 |
96 |
4800 |
4 |
64 |
|
odb-32 (X5-2 Only) |
32 |
64 |
128 |
6400 |
4 |
64 |
|
odb-36 (X5-2 Only) |
36 |
64 |
128 |
7200 |
4 |
64 |
Table B-4 Oracle Database Appliance In-Memory Database Template Size
| Template | CPU Cores | SGA (GB) | PGA (GB) | In-Memory (GB) | Processes | Redo log file size (GB) | Log buffer (MB) |
|---|---|---|---|---|---|---|---|
|
odb-01s (X5-2, X4-2, X3-2, and V1) |
1 |
2 |
1 |
1 |
200 |
1 |
16 |
|
odb-01 (X5-2, X4-2, X3-2, and V1) |
1 |
4 |
2 |
2 |
200 |
1 |
16 |
|
odb-02 (X5-2, X4-2, X3-2, and V1) |
2 |
8 |
4 |
4 |
400 |
1 |
16 |
|
odb-04 (X5-2, X4-2, X3-2, and V1) |
4 |
16 |
8 |
8 |
800 |
1 |
32 |
|
odb-06 (X5-2, X4-2, X3-2, and V1) |
6 |
24 |
12 |
12 |
1200 |
2 |
64 |
|
odb-12 (X5-2, X4-2, X3-2, and V1) |
12 |
48 |
24 |
24 |
2400 |
4 |
64 |
|
odb-16 (X5-2, X4-2, X3-2 Only) |
16 |
64 |
32 |
32 |
3200 |
4 |
64 |
|
odb-24 (X5-2, X4-2 Only) |
24 |
96 |
48 |
48 |
4800 |
4 |
64 |
|
odb-32 (X5-2 Only) |
32 |
128 |
64 |
64 |
6400 |
4 |
64 |
|
odb-36 (X5-2 Only) |
36 |
128 |
64 |
64 |
7200 |
4 |
64 |
Table B-5 Oracle Database Appliance X5-2 Database Template Size Storage Performance
| Template | Number of databases that you can deploy using this template | Container Database ACFS Size on DATA Disk Group (GB) | I/Ops with single storage shelf (HDD/FLASH) | Throughput (MBps) with single storage shelf (HDD/FLASH) | I/Ops with storage shelf plus storage expansion shelf (HDD/FLASH) | Throughput (MBps) with storage shelf plus storage expansion shelf (HDD/FLASH) | Log generation (MBps) |
|---|---|---|---|---|---|---|---|
|
odb-01s |
36 |
100 |
42/4167 |
83/83 |
83/8333 |
167/167 |
6.83 |
|
odb-01 |
36 |
100 |
42/4167 |
83/83 |
83/8333 |
167/167 |
6.83 |
|
odb-02 |
18 |
200 |
83/8.3K |
167/167 |
167/16.7K |
333/333 |
6.83 |
|
odb-04 |
9 |
400 |
167/16.7K |
333/333 |
333/33.3K |
667/667 |
13.65 |
|
odb-06 |
6 |
800 |
250/25K |
500/500 |
500/50K |
1000/1000 |
27.3 |
|
odb-12 |
3 |
1600 |
500/50K |
1000/1000 |
1000/100K |
2000/2000 |
27.3 |
|
odb-16 |
2 |
1600 |
750/75K |
1500/1500 |
1500/150K |
3000/3000 |
27.3 |
|
odb-24 |
1 |
1600 |
1500/150K |
3000/3000 |
3000/300K |
6000/6000 |
27.3 |
|
odb-32 |
1 |
1600 |
1500/150K |
3000/3000 |
3000/300K |
6000/6000 |
27.3 |
|
odb-36 |
1 |
1600 |
1500/150K |
3000/3000 |
3000/300K |
6000/6000 |
27.3 |
Note:
The number of databases that you can deploy using this template is for the entire Oracle Database Appliance.
Actual I/O per second (I/Ops) and throughput for a database depends on the percentage of I/O operations that are serviced through the FLASH, compared with the operations serviced through the hard disk drives (HDDs). The I/Ops weighting in these tables assumes an OLTP workload calculated as a mixed workload (80 percent reads, serviced through the FLASH cache, and 20 percent writes serviced through the HDDs).
Table B-6 Oracle Database Appliance X4-2 Database Template Size Storage Performance
| Template | Number of databases that you can deploy using this template | Container Database ACFS Size on DATA Disk Group (GB) | I/Ops with single storage shelf | Throughput (MBps) with single storage shelf | I/Ops with storage shelf plus storage expansion shelf | Throughput (MBps) with storage shelf plus storage expansion shelf | Log generation (MBps) |
|---|---|---|---|---|---|---|---|
|
odb-01s |
24 |
100 |
138 |
146 |
275 |
229 |
6.83 |
|
odb-01 |
24 |
100 |
138 |
146 |
275 |
230 |
6.83 |
|
odb-02 |
12 |
200 |
275 |
292 |
550 |
458 |
6.83 |
|
odb-04 |
6 |
400 |
550 |
583 |
1100 |
917 |
13.65 |
|
odb-06 |
4 |
800 |
825 |
875 |
1650 |
1375 |
27.3 |
|
odb-12 |
2 |
1600 |
1650 |
1750 |
3300 |
2750 |
27.3 |
|
odb-16 |
1 |
1600 |
3300 |
3500 |
6600 |
5500 |
27.3 |
|
odb-24 |
1 |
1600 |
3300 |
3500 |
6600 |
5500 |
27.3 |
Table B-7 Oracle Database Appliance X3-2 Database Template Size Storage Performance
| Template | Number of databases that you can deploy using this template | Container Database ACFS Size on DATA Disk Group (GB) | I/Ops with single storage shelf | Throughput (MBps) with single storage shelf | I/Ops with storage shelf plus storage expansion shelf | Throughput (MBps) with storage shelf plus storage expansion shelf | Log generation (MBps) |
|---|---|---|---|---|---|---|---|
|
odb-01s |
16 |
100 |
206 |
219 |
413 |
344 |
6.83 |
|
odb-01 |
16 |
100 |
206 |
219 |
413 |
344 |
6.83 |
|
odb-02 |
8 |
200 |
413 |
438 |
825 |
688 |
6.83 |
|
odb-04 |
4 |
400 |
825 |
875 |
1650 |
1375 |
13.65 |
|
odb-06 |
2 |
800 |
1650 |
1750 |
3300 |
2750 |
27.3 |
|
odb-12 |
1 |
1600 |
3300 |
3500 |
6600 |
5500 |
27.3 |
|
odb-16 |
1 |
1600 |
3300 |
3500 |
6600 |
5500 |
27.3 |
Table B-8 Oracle Database Appliance Version 1 Database Template Size Storage Performance
| Template | Number of databases that you can deploy using this template | Container Database ACFS Size on DATA Disk Group (GB) | I/Ops | Throughput (MBps) | Log generation (Mps) |
|---|---|---|---|---|---|
|
odb-01s |
12 |
100 |
333 |
250 |
6.83 |
|
odb-01 |
12 |
100 |
333 |
250 |
6.83 |
|
odb-02 |
6 |
200 |
667 |
500 |
6.83 |
|
odb-04 |
3 |
400 |
1333 |
1000 |
13.65 |
|
odb-06 |
2 |
800 |
2000 |
1500 |
27.3 |
|
odb-12 |
1 |
1600 |
4000 |
3000 |
27.3 |
Note:
For a single database, select a template that best fits your expected workload (the CPU and I/Ops capabilities you require) on the hardware that you are using. When creating multiple databases, calculate the overall workload; available system resources for any new database are affected by the CPU and I/Ops consumed by the existing databases that are already on the system.