[infinispan-dev] Consistent hashes and Hot Rod

[Manik Surtani]

On 30 Aug 2013, at 14:13, Dan Berindei <dan.berindei at gmail.com> wrote:

> We will still have a division in the consistent hash, because the "number of segments" in some ways replaces . We just didn't add it to the HotRod protocol yet...
> 
> I have thought about replacing division, and I was considering 2 options:
> 1. Hide the power of 2 restriction in the consistent hash implementation - allow the user to specify any number of segments, but round it up internally.

^^ This is the technique I've seen in most place.

> 2. Compute the multiplicative inverse of the number of segments and use multiplication instead - multiplication should be just as fast as addition on modern machines.
> 
> However, the division didn't particularly stick out as a hotspot back when I implemented DefaultConsistentHash, so I didn't pursue it further.
> 
> Cheers
> Dan
> 
> 
> On Thu, Aug 29, 2013 at 5:29 PM, Manik Surtani <msurtani at redhat.com> wrote:
> Pls ignore.  I've just been informed that hash space is to be dropped from the protocol anyway.
> 
> On 29 Aug 2013, at 13:35, Manik Surtani <msurtani at redhat.com> wrote:
> 
> > The Hot Rod protocol current passes the size of the hash space used to clients, so that clients are able to create a consistent hash and perform smart routing.
> >
> > https://docs.jboss.org/author/display/ISPN/Hot+Rod+Protocol+-+Version+1.0#HotRodProtocol-Version1.0-HashDistributionAwareClientTopologyChangeHeader
> > https://docs.jboss.org/author/display/ISPN/Hot+Rod+Protocol+-+Version+1.2#HotRodProtocol-Version1.2-HashDistributionAwareClientTopologyChangeHeader
> >
> > However there are no rules as to what this hash space is, and this forces clients to use the remainder operation when performing modular arithmetic.  This has been proven to be inefficient [1] [2] [3], and can be greatly improved if we made a simple assumption that the hash space will always be a power of two.
> >
> > Perhaps something we can add in v1.3 of the protocol, so that clients talking to a server using Hot Rod 1.3 can make the assumption that the hash space is a power of 2?
> >
> > WDYT?
> >
> > - Manik
> >
> >
> > [1] http://dhruba.name/2011/07/12/performance-pattern-modulo-and-powers-of-two/
> > [2] http://scicomp.stackexchange.com/questions/187/why-is-division-so-much-more-complex-than-other-arithmetic-operations
> > [3] http://disruptor.googlecode.com/files/Disruptor-1.0.pdf (Page 5, last paragraph)
> >
> >
> 
> --
> Manik Surtani
> 
> 
> 
> 
> _______________________________________________
> 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

--
Manik Surtani



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.jboss.org/pipermail/infinispan-dev/attachments/20130830/d4564c17/attachment-0001.html