POHMELFS userspace server and the latest to date (pushed into the git) kernel client.
In-kernel async NFS server and client (2.6.28 tree).
Both client and server have 8Gb of RAM, 32bit CPUs.
The first bench: untarring the linux kernel sources (smaller is better):
POHMELFS: 5 seconds NFS: 44 seconds
Sync called after the untar takes less than a second, so it does not affect the test.
The second one: dbench with 30 threads. Servers exported /dev/shm to the clients.
Updated POHMELFS version fixes fair number of bugs both from its move from own path to dentry cache and old cache coherency issues. There is a nasty issue in the code: object removal. It is not optimized at all, and, hmm, it does not work for directories :) It happens because unhashed direntry path lookup in the dcache returns not only path to the direntry, but also adds (deleted) string after it, so obviously object with such a pathname can not be removed.
I will work on this before pushing upstream (scheduled for tomorrow).
Stay tuned!
yuhuuuu upstream :)
Mmm... Dounuts! ;)
One-man very active upstream, you mean? Or... ah-ah, yeah, "please do/submit/go" upstream... Didn't you read "that e-mail"? :-P http://www.ioremap.net/node/72#comment-154 http://www.ioremap.net/node/120 Ehrm.. no %) , that one. http://www.ioremap.net/node/66
2zbr: Good job making fun and $) publicity with the projects. Crazy idea, make an NFS-server front-end to the *FS, make it feature-bug-testcase(is there one?) compatible with and benchmark on the "same ground"? Then go upstream and replace %) the predecessor. World Domination is fun, can you see? :))))))