HAMMER2 and clustered writing

Matthew Dillon (re?)added a sysctl: vfs.hammer2.cluster_write.  It defaults to off, since HAMMER2 already writes a large buffer size and this should, in theory, not be needed.  It may improve performance in some situations where there’s a lot of file creation and deletion, but that’s my theoretical guess rather than anything I’ve bennchmarked.

2 Replies to “HAMMER2 and clustered writing”

  1. When I first read the title, I was hoping for multi-master writing in a cluster of dragonfly machines.

  2. Hmm, this seems to be sort of like ZFS’s write cache. DragonFly has had a read-cache (swapcache) for some time now. This I imagine should also make writes to heavily written to databases go more smoothly.

Comments are closed.