[infinispan-dev] TotalOrder and jgroups

[Bela Ban]

On 3/6/12 11:34 PM, Mircea Markus wrote:
>> I say our best bet for now is to communicate with all stakeholders
>> (e.g.
>> Paul for the AS), and come up with agreed-upon sets of configs, to be
>> used perhaps over a shared transport. I reconsider my previous
>> statement
>> and now think it's best to avoid hidden insertions of protocols.
>
> +1 - I've started doing that and it's an ugly business.
>
> Here's another approach:
> - all jgroups configs shipped with ISPN come with the SEQUENCER added
> - by default all messages have NO_TOTAL_ORDER flag attached
> - RpcManager.invoke mthods would also support a totalOrder flag (similar to OBB flag currently supported)
>
> With this approach:
> - we would support heterogeneous caches on the same cache manager: TO, 2PC *and* non-tx caches
> - we would allow existing users to switch to total order replication without the need to change the jgroups configuration
> - the performance penalty of having the SEQUENCER for caches that don't need it, e.g. not tx caches, is minimal by the look of the code
>
> How does it sound?

Intuitively, not nice ! Can't we have different configurations for these 
use cases ?

-- 
Bela Ban
Lead JGroups (http://www.jgroups.org)
JBoss / Red Hat