[erlang-questions] Distributed Process Registry

Ulf Wiger <>
Mon Feb 9 00:04:08 CET 2015


> On 08 Feb 2015, at 23:07, Roberto Ostinelli <> wrote:
> 
> I'm actually very ok with the AP portion of the CAP theorem (that pg2 seems to offer) in contrast with the CA (of global/gproc).

A minor detail: When recovering from netsplits in gproc, only unique names must be resolved so that there is only one (by definition of ‘unique’).

If you want to implement a global broadcast group, you would do that using ‘properties’, and they are simply merged after netsplit - there is no uniqueness constraint.

BR,
Ulf W

Ulf Wiger, Co-founder & Developer Advocate, Feuerlabs Inc.
http://feuerlabs.com






More information about the erlang-questions mailing list