cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ariel Weisberg (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-13442) Support a means of strongly consistent highly available replication with storage requirements approximating RF=2
Date Tue, 18 Apr 2017 04:19:41 GMT


Ariel Weisberg commented on CASSANDRA-13442:

bq. if you have three copies and you lose one, no big deal, you still have two to restore
from. Just two copies? If anything goes wrong with that other copy while you repair you are
I just realized you are getting thrown off by RF=3 N=1. That is just an example. You can run
RF=6 N=2 with 3 replicas in each data center and read/write at LOCAL_QUORUM. You still save
1/3 storage.

A lot of use cases are massively over replicated when it comes to simply preserving data from
hardware failure. Ideally we could use erasure coding, but you still need to make strong consistency
happen. I think part of the issue is that there is a conflation of how we determine what the
state should be with how we replicate the state.

> Support a means of strongly consistent highly available replication with storage requirements
approximating RF=2
> ----------------------------------------------------------------------------------------------------------------
>                 Key: CASSANDRA-13442
>                 URL:
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Compaction, Coordination, Distributed Metadata, Local Write-Read
>            Reporter: Ariel Weisberg
> Replication factors like RF=2 can't provide strong consistency and availability because
if a single node is lost it's impossible to reach a quorum of replicas. Stepping up to RF=3
will allow you to lose a node and still achieve quorum for reads and writes, but requires
committing additional storage.
> The requirement of a quorum for writes/reads doesn't seem to be something that can be
relaxed without additional constraints on queries, but it seems like it should be possible
to relax the requirement that 3 full copies of the entire data set are kept. What is actually
required is a covering data set for the range and we should be able to achieve a covering
data set and high availability without having three full copies. 
> After a repair we know that some subset of the data set is fully replicated. At that
point we don't have to read from a quorum of nodes for the repaired data. It is sufficient
to read from a single node for the repaired data and a quorum of nodes for the unrepaired
> One way to exploit this would be to have N replicas, say the last N replicas (where N
varies with RF) in the preference list, delete all repaired data after a repair completes.
Subsequent quorum reads will be able to retrieve the repaired data from any of the two full
replicas and the unrepaired data from a quorum read of any replica including the "transient"

This message was sent by Atlassian JIRA

View raw message