Hi, While doing a quick search for this mailing list I found that there's been some work done (recently) in adding a writeback cache for NFS ontop of fscache (here: http://www.jiayi.net/works/nfsfscache-writeback-cache/#NFS_1) Has this been merged previously? It looks like the guy was able to get some impressive write performance improvements. I ask because I would love to integrate it into the fscache code for Ceph when I'm fairly confident in the RO implementation. Our use case involves a very large distributed database for analytics data. Our machines that end up performing the crunching of the data have large SSD drives that we're planning on using as a local cache for the Ceph filesystem. Being able to do write-back would greatly help with our cases when we're bulk importing new data / merging small table segments. Best, - Milosz -- Linux-cachefs mailing list Linux-cachefs@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cachefs
