incubator-cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jonathan Ellis <jbel...@gmail.com>
Subject Re: OrderPreservingPartitioner limits and workarounds
Date Wed, 07 Apr 2010 19:13:13 GMT
One thing you can do is manually "randomize" keys for any CFs that
don't need the OP by pre-pending their md5 to the key you send
Cassandra.  (This is all RP is doing under the hood anyway.)

On Wed, Apr 7, 2010 at 5:51 AM, Paul Prescod <paul@prescod.net> wrote:
> I have one append-oriented workload and I would like to know if
> Cassandra is appropriate for it.
>
> Given:
>
>  * 100 nodes
>
>  * an OrderPreservingPartitioner
>
>  * a replication factor of "3"
>
>  * a write-pattern of "always append"
>
>  * a strong requirement for range queries
>
> My understanding is that there will exist 3 nodes will end up being
> responsible for all writes and potentially a disproportionate amount
> of reads (in the common case that users care more about recent data
> than older data).
>
> Is there some manual way I can fiddle with InitialTokens and
> ReplicationFactors to share the load more fairly?
>
>  Paul Prescod
>

Mime
View raw message