PRE-REQUISITES
1) Stop user activity and backup data
Follow standard procedures.
The objective is to make sure that data are not accessed not updated.
This means an interruption of service. Duration can vary according to volume of data.
Dummy run will enable to evaluate real duration.
2) Verify that a physical backup of data is available
Follow standard procedures.
3) Dispatch of delete all pending workspaces
Follow standard procedures.
The objective is that no pending update exist.
They can be exported as MGR file, cancelled or dispatched.
3) Verify you have enough working space
The process used the reorganization feature of HOPEX.
For each repository, data will be dumped to a large command file on disk using logical backup feature.
The generation folder is a subfolder of HOPEX environment folder. There is one file per repository. Ex:
<Myenvironment path>\SysDb\WORK
<Myenvironment path>\Db\MyDataRepository\WORK
4) Create SQL Server database with appropriate permissions
See SQL Server DBA.
For each repository (data repository or system repository) a SQL Server is required.
SQL Server database must be empty.
A SQL server user with appropriate permission must be created.
See HOPEX Administrator for details.
For more details, see only documentation.
Installation and Deployment : RDBMS Repository Installation Guide : SQL Server support, SQL Server Requirements.
Basic metrics:
For logical backup file: size of Oracle dump for the schema of the repository x 5
For logical backup file: size of Oracle dump for the schema of the repository x 1
|
Example
|
Size of .DMP in Oracle
|
Size expected for SQL Server database
|
Disk expected for logical backup file
|
|
Schema of MyRepository
|
5 Go
|
5 Go
|
25 Go
|
|
Schema of SystemDb
|
3 Go
|
3 Go
|
15 Go
|
5) Performances tuning indications
The article below contains optional indications to favor performances when performing mass import in SQL Server