[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [wsrm] Preliminary minutes of WSRM TC Conf call -050603
Paolo Romano wrote: > >4.4 REL-11 MESSAGE ID AND GROUPID/SEQUENCENO > .. > >Paulo stated that having two identifiers (such as groupID, sequenceNO) is > >needed for several groups of messages with dependencies. Messages ordered > >by groups is a use case we have often. He also would prefer one id > >mechanism, but it could be a struct with groupID > > I guess I must have missed a "not" in the yesterday phone call... > Actually, I can't imagine many common use cases where between two end > points it is necessary to have simoultaneously multiple groups of messages I think this is a common use case. Take the case of Sender A processing purchase orders for 2 different customers ('x' & 'y) simultaneously. It likes to talk to Receiver B (a Warehouse company) in different steps, say, first like do a inventory check, then a credit check, and finally an order. Sender would like to create a group for each customer and hence there could be 2 simultaneous group exchanges between node A and node B. That said, I agree that we should avoid having 2 different primary keys. However, I'm not in favor of Doug's proposal on the call regarding "depends on" concept. The problems I see there are: i) RM processors will act then as a mini Workflow processors - we don't want to get into workflow stuff :) ii) Could potentially have cyclical dependencies if not spec-ed out properly. iii) Concept of one universal group preempts vendors doing implementation optimizations such as removing the message content (not the metadata) from the persistence store once the 'group' is complete. Some random thoughts how to achieve this are: i) Get rid of MId and have GId/Sid set. Gid could be optional & SId is mandatory. If Gid is present, SId is the unique in that GId. If not, SId should be globally unique. ii) Or, always mandate GId/SId and have a different semantics for GId '0'. GId '0' should be considered un-ordered sequences. Thoughts??? -Sunil > > which have to be ordered independently. This is why I consider redundant > having two different identifiers (Group_id + Sequence_id), and I would > prefer to use only one identifier for both duplicate elimination and > > sequence ordering. The advantages of such an approach would be: > implementation simplification and reduced "verbosity" => overhead. > > May be somebody can show us some use cases where the feature of having > multiple groups of messages ordered is worthy the cost of using distinct > identifiers for ordering and filtering out duplicates. > > Paolo
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]