Re: Riak doesn't use consistent hashing.

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Riak doesn't use consistent hashing.

Justin Sheehy
Hi, Greg.

Thanks for your thoughtful analysis and the pull request.

On Thu, May 26, 2011 at 1:54 AM, Greg Nelson <[hidden email]> wrote:

> However, the skipping bit isn't part of
> Riak's preflist calculation.  Instead, nodes claim partitions in such a way
> as to be spaced out by target_n_val, to obviate the need for skipping.

A fun bit of history here: once upon a time, Riak's claiming worked in
the same way as described by Amazon, with "skipping" and all.  We
noticed that this approach caused a different set of operational
difficulties when hinted handoff due to node outages was occurring at
the same time as a membership change.  That prompted changes to the
claim algorithm, which we still consider an area deserving of active
improvement.

Multiple people will be reading, analyzing, and testing your work to
contribute to this improvement.  We very much appreciate your efforts,
and want to make sure that we incorporate them in the best possible
way.

Thanks,

-Justin

_______________________________________________
riak-users mailing list
[hidden email]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
Reply | Threaded
Open this post in threaded view
|

Re: Riak doesn't use consistent hashing.

Greg Nelson
Excellent.  Let me know what I can do to help.

-Greg

On Thursday, May 26, 2011 at 7:37 AM, Justin Sheehy wrote:

Hi, Greg.

Thanks for your thoughtful analysis and the pull request.

On Thu, May 26, 2011 at 1:54 AM, Greg Nelson <[hidden email]> wrote:

 However, the skipping bit isn't part of
Riak's preflist calculation.  Instead, nodes claim partitions in such a way
as to be spaced out by target_n_val, to obviate the need for skipping.

A fun bit of history here: once upon a time, Riak's claiming worked in
the same way as described by Amazon, with "skipping" and all. We
noticed that this approach caused a different set of operational
difficulties when hinted handoff due to node outages was occurring at
the same time as a membership change. That prompted changes to the
claim algorithm, which we still consider an area deserving of active
improvement.

Multiple people will be reading, analyzing, and testing your work to
contribute to this improvement. We very much appreciate your efforts,
and want to make sure that we incorporate them in the best possible
way.

Thanks,

-Justin


_______________________________________________
riak-users mailing list
[hidden email]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com