If you are asking are your details correct for what I described, I don’t think so but I have a feeling I don’t completely understand you situation.
I’m just describing our general technique for handling bottleneck or problematic interfaces in general.
the idea is to move the bottle neck part of the integration into its own process and or site to keep its bad behaivor from impacting those interfaces that are playing nicely in the sand box or process in our case.
It would be more like this:
ib_lab
PROTOCOL: pdl-tcpip
Type: Server
Port: 6040
process_1
raw route
hs_12345
PROTOCOL: pdl-tcpip
Type: Client
Host: localhost
Port: 12345
process_1
——————————
hr_12345
PROTOCOL: pdl-tcpip
Type: Server
Port: 12345
process_2
xlated route
ob_hospital
PROTOCOL: pdl-tcpip
Type: Client
Host: 172.16.10.31
Port: 7050
process_2
Russ Ross
RussRoss318@gmail.com