Re: IBM i in the cloud (was social media)

On 11/22/2010 4:05 PM, Nathan Andelin wrote:

Could we explore that idea a little further? And just to reiterate a little bit
of background to keep things in context, we're talking about using a single IBM
i box to handle requests for multiple tenants who each have their own separate
database collection (library). So we're exploring the idea of writing a
connection pool manager to return a connection to a DB server instance which
would use the appropriate library list for the requesting tenant.

It's getting late and I must be getting confused, because AFAIK a connection pool only returns a connection which matches the parameters you are requesting. Those parameters typically include user ID, password, and the connection URL, which in turn contains the connection properties. The properties can be used to set up, for example, the default schema. So in this case a connection would by definition only return a connection for the correct tenant. The work on your part is to identify the correct connection string for each tenant.

The premise also cuts off some plausible alternatives. There absolutely is no one-size-fits-all strategy. For example, persistent connections don't need any of this. The connection sets its library list at the time it is created and you use unqualified table names and system naming.

Joe