I’m not going to follow the detailed format that I’ve seen some of the other teams using, because I’m lazy, but here’s a brief snapshot of where NSR development is now and where it’s going.
*** Design Info Avra has pulled together pieces from various other documents into a spec here: https://docs.google.com/document/d/1bbxwjUmKNhA08wTmqJGkVd_KNCyaAMhpzx4dswokyyA/edit?usp=sharing He has also produced some separate UML diagrams: https://docs.google.com/presentation/d/1lxwox72n6ovfOwzmdlNCZBJ5vQcCaONvZva0aLWKUqk/edit?usp=sharing If you’re interested specifically in the journal translator (which has uses beyond NSR), there’s slightly more information here: http://pl.atyp.us/misc/journal-xlator.pdf *** Code Several necessary pieces of infrastructure, such as GF_FOP_IPC and the basic code generator, have already been merged separately. Both the I/O path code and journaling code are very much in progress. Some of the relevant patches are here: http://review.gluster.org/12388 (client side) http://review.gluster.org/12705 (server side) http://review.gluster.org/12450 (journaling) http://review.gluster.org/8887 (etcd support) *** Near Term Plans The immediate near-term plan is to get the above pieces completed so that we can have a demo-able version of the I/O path next month. This will also allow us to measure performance overhead. There are still many more improvements that will need to be made, especially to journaling, plus we need to revive some of the old reconciliation code from last year. That plus adding tests will represent the major efforts for the next few months at least. _______________________________________________ Gluster-devel mailing list [email protected] http://www.gluster.org/mailman/listinfo/gluster-devel
