Hi, I already introduced the current outcome of my work on Ceph deduplication on the last CDM, but I thought it might be a good idea to do so on the list as well :). tl;dr: There is now a client tool / library to use a Ceph pool as content-addressed storage. It writes data with object names equal to the data's hash. Two times the same data therefore yields only one object with a reference count of 2. More detailed design description and documentation is on github: https://github.com/irq0/veintidos Unfortunately / fortunatly I can't continue to work on bringing the prototype to C++ or improving it otherwise, since i'll go on an extended vacation till end of this year pretty soon.. In the long run, I think, it could be nice for archives ( iRODS?) or as an alternative to the Striper in RGW. Please let me know what you think, ~marcel -- Marcel Lauhoff Mail: lauhoff@xxxxxxxxxxxx XMPP: mlauhoff@xxxxxxxxxxxxxxxxxxx -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
