![]() |
|
|
Welcome to the { mindfrost82.com } forums. You are currently viewing our boards as a guest which gives you limited access to view most discussions and access our other features. By joining our free community you will have access to post topics, communicate privately with other members (PM), respond to polls, upload content and access many other special features. Registration is fast, simple and absolutely free so please, join our community today! If you have any problems with the registration process or your account login, please contact contact us. |
|
|||||||
![]() |
|
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
|
|||
|
XP - NFS Inode Cache Performance in 2.6 Kernels
I read a kerneltrap post (http://kerneltrap.org/node/4462) stating the
NFS Inode cache in the 2.4 kernel has performance issues when it gets too full it causes a great deal of hash collisions, and is a performance issue. I briefly traced the inode lookup code in the 2.4 kernel and found that it seems to have a number of hash buckets, and perform external chaining to handle collisions. I'm using a 2.6 kernel, and performing a high-performance parallelized directory walk, resulting in my kernel's slab allocator using 3GB of RAM for the NFS inode cache, and the dentry cache. I have some concerns that I might be affected by a similar performance problem. I briefly traced the 2.6 kernel's inode lookup function, and found that it seems to use a proper hashtable (but I neither have personal knowledge of the table's algorithm, nor do I have time, unfortunately, to investigate). If anyone has any information on this topic cached, I'd be appreciative if it could be shared. thanks, Friendly Anonymous Poster |
![]() |
|
| Thread Tools | Search this Thread |
| Display Modes | |
|
|