[infinispan-dev] asynchronous change manager for high load Lucene

[Manik Surtani]

On 30 Mar 2010, at 13:00, Emmanuel Bernard wrote:

> 
> On 30 mars 2010, at 13:14, Sanne Grinovero wrote:
> 
>> 
>> Emmanuel Bernard wrote:
>>> Bela Ban wrote:
>>>> You could use the cluster view, which is the same across all nodes, and
>>>> pick the first element in the list. Or you could run an agreement
>>>> protocol, which deterministically elects a master.
>>> 
>>> Looks simple, deterministic and elegant.
>> 
>> Agreed, that looks like a clean solution, but would scale better if we
>> could decide a master for each different index, to spread the load
>> instead of choosing always the same node for all the work, ideas on
>> that?
>> I could hash the indexname (the identifier) and % cluster view list size, WDYT ?
> 
> That would work assuming you take the actual directory provider name (ie the one that can potentially contain the sharded information like Account.1, Account.2 etc).

This sounds very workable, yes.

--
Manik Surtani
manik at jboss.org
Lead, Infinispan
Lead, JBoss Cache
http://www.infinispan.org
http://www.jbosscache.org