Hello
I would like some advice on the best way to handle mutliple HFCS databases on a single server.
We have an MS multi-tenanted server that serves 20 to 25 clients.
Each client has their own separate HFCS database. The users per client vary from 1 to 10 so we have a mix of light and heavy use clients on each server. The data in the light use databases is about 1GB and there can be 4-6GB in the heavy use clients. The Windev app is re-entrant so a single user may have several copies of the app running. There never seems to be more than 30 users logged in at any one time. At the moment there is a single manta service running and it hosts the 20-25 databases.
One problem in this approach is that I cannot restart a single database because I have to restart the Manta server, which then shuts down all the databases. I also wonder about running that many databases on a single Manta server - is that good practice ?
One solution I can see is to run mutliple Manta servers - one per database - but that means that the server is then running 20-25 Manta servers with all the associated overheads and splitting the memory between them all.
Regards
Al
I would like some advice on the best way to handle mutliple HFCS databases on a single server.
We have an MS multi-tenanted server that serves 20 to 25 clients.
Each client has their own separate HFCS database. The users per client vary from 1 to 10 so we have a mix of light and heavy use clients on each server. The data in the light use databases is about 1GB and there can be 4-6GB in the heavy use clients. The Windev app is re-entrant so a single user may have several copies of the app running. There never seems to be more than 30 users logged in at any one time. At the moment there is a single manta service running and it hosts the 20-25 databases.
One problem in this approach is that I cannot restart a single database because I have to restart the Manta server, which then shuts down all the databases. I also wonder about running that many databases on a single Manta server - is that good practice ?
One solution I can see is to run mutliple Manta servers - one per database - but that means that the server is then running 20-25 Manta servers with all the associated overheads and splitting the memory between them all.
Regards
Al