[infinispan-dev] Shipping logical address String in Hot Rod protocol v1.1?

[Galder Zamarreño]

On Nov 30, 2011, at 10:28 AM, Dan Berindei wrote:

> On Wed, Nov 30, 2011 at 10:04 AM, Bela Ban <bban at redhat.com> wrote:
>> 
>> 
>> On 11/30/11 8:59 AM, Dan Berindei wrote:
>> 
>>>> Why don't you send the UUID as a (16 byte) string then ?
>>>> 
>>> 
>>> Yeah, that would work. However, a UUID is not always a valid UTF-8
>>> string, so we should probably define it in the protocol as an array of
>>> bytes (without any meaning).
>> 
>> 
>> Yes. We did something similar in JGroups, take a look at
>> ENCRYPT.byteArrayToHexString().
>> 
> 
> Since HotRod is a binary protocol I meant to send the UUID as it is
> (no encoding whatsoever).
> 
> I think sending the UUID as a hex-encoded string, a raw byte array, or
> a 32-bit hash (since that's all we need on the client) are all valid
> choices.

Right, I think relying on the UUID underneath the JGroups address is a better choice than the logical address itself, so let's use that as base.

My preference is for a 32-bit hash which is nice and compact.

https://issues.jboss.org/browse/ISPN-1578

> 
>> --
>> Bela Ban
>> Lead JGroups (http://www.jgroups.org)
>> JBoss / Red Hat
>> _______________________________________________
>> infinispan-dev mailing list
>> infinispan-dev at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/infinispan-dev
> _______________________________________________
> infinispan-dev mailing list
> infinispan-dev at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/infinispan-dev

--
Galder Zamarreño
Sr. Software Engineer
Infinispan, JBoss Cache