Re: Bind by reference doesn't mean dynamic call?

David,

Having a problem explaining to colleagues in a world without service
programs that using them does not mean dynamic calls and thus poor
performance.

I think it's a mistake to try to sell service programs as a technology based on performance. The performance difference between a dynamic call and a bound-by-reference call is so slight that it's never likely to matter.

If I understand correctly, the first call is akin to being dynamic,
but afterwards the link between program and service program is static
and therefore as fast as a module bound by copy.

This seems to be discussing whether the program is reactivated on each call or not. That doesn't depend on whether it's a dynamic or static call, instead it depends on how the program is activated.

In ILE, a program remains activated until it's activation group is reclaimed (either explicitly or automatically.) In OPM it remains activated until it ends with LR turned on.

Programs that remain activated will call very quickly, as they're already loaded into memory, already activated, may still have their files open, etc, etc.

Anyone help me out in simple English?

In simple English:

Advantage to ILE is maintainability, not performance.