The ability to insert multiple database rows in a single SQL insert statement. The ANSI syntax for multi
row is shown in
Comparing single row and multi row insert syntax.
Table 1. Persistence API configuration properties
| Category |
Key |
Valid Values |
Default |
Comment |
| CORE |
MRI_ROW_CACHE_LIMIT |
+ve Integer |
1 |
Sensible upper limits may depend on the usecase and the system/application being
used. Ideally set high enough that all inserts in a unit of work can be cached, bearing in mind that this will
drive memory consumption. Expectation that values up to 10,000 are realistic, above this evaluate with
care. |
| CORE |
MRI_ROWS_PER_INSERT_LIMIT |
+ve Integer |
1 |
Currently forced to 1 for z/OS®. Values in the
range 1-500 are probably most sensible. |
| CORE |
MRI_USE_TABLES |
Y | N |
N |
Currently forced to Y for Oracle. |
| CORE |
MRI_EM2_ENABLED |
Y | N |
Y |
Supports a fallback mechanism in case MRI causes problems with a specific inbound
map. |
The FTM Persistence API supports multi row inserts as well as providing the following benefits:
- Abstraction from database specific syntax
- Configurable control of the rows per insert
- Configurable control of the number of rows to cache in memory
- Ability to insert using the 'insertable' views
- Ability to insert direct to the tables
- Full FTM instrumentation of all SQL issued through the API
- Trace of SQL errors including SQL state and error codes