Re: [Gluster-devel] Why vandermonde matrix is used in EC?

On 11/27/2016 09:58 PM, 한우형 wrote: Hi, Thank you so much for the speedy reply, but I have some more questions. 1) I understand Non-systematic encoding/decoding doesn't alter performance when one or more bricks are down. but why systematic approach has service degradation? I think when parity

Re: [Gluster-devel] Merge window closed for the approaching GlusterFS-3.7.18 release

On 11/27/2016 12:34 PM, Samikshan Bairagya wrote: > > On 11/25/2016 05:24 PM, Samikshan Bairagya wrote: >> Hi all, >> >> GlusterFS-3.7.18 is on target to be released on November 30th >> (Wednesday). >> >> In preparation for the release, maintainers would need to merge all >> changes into

Re: [Gluster-devel] Merge window closed for the approaching GlusterFS-3.7.18 release

- Original Message - From: "Samikshan Bairagya" To: "Gluster Devel" , "GlusterFS Maintainers" gluster.org> Sent: Sunday, November 27, 2016 12:34:11 PM Subject: [Gluster-devel] Merge window closed for the approaching GlusterFS-3.7.18

[Gluster-devel] Question about EC locking

Hi Xavier, Notice that EC xlator uses blocking locks. Any specific reason for this? Do you think this will affect the performance ? (In comparison AFR first tries non blocking locks and if not successful, tries blocking locks then) Also, why two locks are needed per FOP ? One for

[Gluster-devel] Question about EC Locking

Hi Xavier, Noticed that EC xlator uses blocking locks. Any specific reason for this ? Do you think this will affect the read/write performance ? ___ Gluster-devel mailing list Gluster-devel@gluster.org

[Gluster-devel] Release 3.10: Cut-off for feature proposal

Hi, We are looking at *Dec. 02' 2016* as the cut-off for any feature proposal for the 3.10 release. So if you have any features that you think would be ready for 3.10, then please do consider opening issues and posting to devl list on the details of the feature, by this date. Thanks, Shyam

Re: [Gluster-devel] Why vandermonde matrix is used in EC?

Hi, Thank you so much for the speedy reply, but I have some more questions. 1) I understand Non-systematic encoding/decoding doesn't alter performance when one or more bricks are down. but why systematic approach has service degradation? I think when parity part is down there's no performance