cassandra-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jeff Jirsa"<>
Subject Re: Guaranteed order for syncing Records
Date Thu, 06 Jul 2017 17:29:44 GMT

On 2017-07-05 23:14 (-0700), suraj pasuparthy <> wrote: 
> thanks Pranay,
> But is the order maintained across tables?
> As in the client in DC1 first writes record A in table A and then
> immediately writes Record B in Table B,
> So will the order be maintained, where in Table A gets synced first and
> then Table B?

Cassandra doesn't use traditional sync mechanisms like you may be assuming (there is no binary
transaction log that is replayed in order on each server). The order of mutations arriving
on the replicas is NOT guaranteed at this time.

If you need to guarantee the above behavior, you may be able to accomplish it with either
sufficiently high consistency levels (block the acknowledgement of the write until N replicas
have responded, including the remote datacenters) at the cost of query latency.

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message