[Gluster-users] How caches are working on AFR?

[Anand Babu Periasamy]

Write-behind is not the right place to do that. It does not even know that volumes
underneath are replicated. Replicate module has options to turn off atomic writes
though.

Instead of making reliability an option, we are making it a must in 2.0, We are working
hard to achieve performance on top of it. 2.0.0RC4 release has some more improvements to
replicate's performance with atomic writes. One more set of improvements will come in RC5
release.

For remote site replication, we are planning on introducing (HSM) Hierarchical storage
management translator, that can sync on close or sync on write or even call an external
tool (that can gzip and rsync). Don't ask for any ETAs now. Our top priority is to make
2.0.0 stable release first :)

--
Anand Babu Periasamy
GPG Key ID: 0x62E15A31
Blog [http://ab.multics.org]
GlusterFS [http://www.gluster.org]
The GNU Operating System [http://www.gnu.org]



Evan wrote:
> Anand Babu Periasamy,
>  
> 
>     On the write-behind translator is there a way to wait for one of the
>     AFR Replica's to get a close response and then finish replicated the
>     data in the background (which write-behind currently does) and issue
>     the close system call to the replica servers long after the
>     application has moved on because at least one of the replicas is
>     keeping up?
> 
>     Thanks
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users