Hi David,
Ok, I don't involve with source code.
Just say we have boss_cache process uses cache application. cache
application has 1 gen_server, it handle GET/SET/DELETE message (gets key,
returns value; set {key, value} entry; delete key).
There are 2 solutions:
1. boss_cache process send message to cache application directly.
2. boss_cache process creates poolboy supervisor, its children are poolboy
workers.
boss_cache sends mesages to poolboy, it checks out worker. The worker sends
message to cache application. Poolboy checks out the worker.
CB is using (2) solution. Is (2) better than (1) ?
For asynchronous SET/DELETE message, obviously we don't need poolboy.
For synchronous GET message, we send message with timeout.
As Even mentioned in
roadmap<https://github.com/ChicagoBoss/ChicagoBoss/wiki/Roadmap#wiki-goal-5-performancescalability-testing>,
he uses poolbooy to resolve bottleneck issue. Maybe I don't understand
poolboy implementation ?
Regards,
Cuong Th.
--
You received this message because you are subscribed to the Google Groups
"ChicagoBoss" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
Visit this group at http://groups.google.com/group/chicagoboss.
To view this discussion on the web visit
https://groups.google.com/d/msgid/chicagoboss/66bc67f3-a41f-40ef-b5bb-a236e0da9373%40googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
