Thanks. what's more, I want to know if there is best practice for
performance in this special situation, such as avoiding to do
something, etc.

 On Wed, Jul 11, 2012 at 11:03 AM, Sean Cribbs <[email protected]> wrote:
> Note he said "secondary index", and yes installing the search hook on that
> many buckets would clog up the ring state.
>
>
> On Tue, Jul 10, 2012 at 10:58 PM, Andrew Zeneski <[email protected]>
> wrote:
>>
>> Sean,
>>
>> Doesn't enabling search on a bucket mean setting properties? At least the
>> precommit hook property right?
>>
>> Andrew
>>
>> On Jul 10, 2012, at 10:56 PM, Sean Cribbs wrote:
>>
>> Wu,
>>
>> As long as you are not setting properties on those buckets, there are very
>> few reasons why performance would be adversely affected. The bucket can
>> simply be considered a prefix for the key.
>>
>> On Tue, Jul 10, 2012 at 10:52 PM, Wu Ray <[email protected]> wrote:
>>>
>>> Hi, guys
>>>
>>> The common situation is there is hundreds buckets, every bucket has
>>> millions of kv pairs. But my question is that I want millions of
>>> buckets , every bucket has kv pairs from dozens to hundreds.
>>>
>>> So, in this condition, is Riak stiil a good solution, I means the
>>> performance and second index search.
>>>
>>> Best regards.
>>>
>>> _______________________________________________

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

Reply via email to