The HPE Distributed Action Handler (HPE DAH) is a server that distributes ACI actions to HPE IDOL Servers. Distribution enables you to scale your system in a linear manner, increasing the speed at which it runs actions and saving processing time. Having multiple IDOL Servers also ensures uninterrupted service if any IDOL Server fails.
The setup of your IDOL Servers is independent of the HPE DAH, because they are architecturally unaware of it.
You can run the HPE DAH in the following modes:
Mirror mode. The child servers (the servers that the HPE DAH distributes actions to) are identical—that is, each one is configured the same way and contains the same data as the others.
Non-mirror mode. The child servers are all different—that is, each one is configured differently and contains different data from the others. If you run the HPE DAH in non-mirror mode, you must set up virtual databases (VDBs) of the following types:
Combinator VDB. The virtual database forwards an action to all the databases that it represents. The VDB collates and sorts the results before it returns them.
Distributor VDB. The virtual database forwards an action to one of the databases that it represents. These databases must be identical (that is, all the databases are exact copies of each other and contain the same data). The distribution method determines which database the distributor VDB forwards actions to.
Virtual databases can map to IDOL Server databases or to other virtual databases that you set up for the HPE DAH.
However, in general, HPE recommends that combinator virtual databases map only to other virtual databases (combinator VDBs combine results from distributor VDBs). Distributor databases normally map directly to the IDOL Server databases. HPE recommends that you do not use them to distribute queries to combinator VDBs.
|