| Author |
Message
|
| RocknRambo |
 Posted: Mon Nov 10, 2008 6:29 am Post subject: DB connectivity |
 |
|
Partisan
Joined: 24 Sep 2003
Posts: 355
|
how does the connectivity to a database established either for compute or db nodes ?
If multiple compute nodes are connected to the same application database (eg: SQL Server), does each compute create its own db thread or the same thread shared across? Is there a concept of db pooling in MB?
any pointing is appreciated.
-RR |
|
| Back to top |
|
 |
| zpat |
| Posted: Mon Nov 10, 2008 6:33 am Post subject: |
|
|
Jedi Council
Joined: 19 May 2001
Posts: 5867
Location: UK
|
| Multiple compute nodes in the same or different message flows? |
|
| Back to top |
|
|
| sridhsri |
| Posted: Mon Nov 10, 2008 6:33 am Post subject: |
|
|
Master
Joined: 19 Jun 2008
Posts: 297
|
Most nodes use ODBC to connect with the DB - exceptions are JCN (can optionally use JDBC) and DataRetrieve node (JDBC).
Any connection pooling done by MB is not available for administration as is with in the case of most Application Servers (so even if it does do any poling you cannot change it or view it).
I would presume that each thread would have one connection with one database. I can say for sure that compute nodes do not start any thread of their own. |
|
| Back to top |
|
|
| RocknRambo |
| Posted: Mon Nov 10, 2008 6:40 am Post subject: |
|
|
Partisan
Joined: 24 Sep 2003
Posts: 355
|
Multiple message flows with compute or db nodes. does it mean each message flow creates a db thread or the execution grp creates the thread?
As a best practice, how is the DNS name (alias on odbc.ini) created? in the scenario of multiple message flows trying access same the application database for different tables or so. |
|
| Back to top |
|
|
| PeterPotkay |
| Posted: Mon Nov 10, 2008 7:17 am Post subject: |
|
|
Poobah
Joined: 15 May 2001
Posts: 7723
|
|
| Back to top |
|
|
|
|
