Mitch Collinsworth wrote: > > On Fri, 3 Feb 2006, Jeffrey Altman wrote: > >> People talk about how bad the performance of AFS is. It is nowhere near >> as bad as the performance of SMB. > > Do you have numbers to back this up? I don't but having them would be > helpful. Probably to many more sites than just ours. The last time > I remember hearing numbers (a while ago now) it was the other way around.
I don't have hard numbers. Its a simple reality. If you are using an AFS cache manager with anything close to a decent cache size, if you need to access a file more than once, then the second time you access it assuming it has not changed the reads will occur a local disk speed instead of WAN speed. In my office I must use SMB to grep the web server logs on a machine sitting 10 feet away over 100Mb/sec networking. Each pass over the logs is painful compared to copying the logs to local disk and then performing multiple passes. It does depend on your environment. If you are crossing networks with packet loss, AFS is impacted more severely than it needs to be. This may be due to a bug in RX/UDP that I have not tracked down yet. In that case SMB may be better because it does not have the need to resync after each RPC as it is communicating over TCP. When the RX/TCP work is complete, AFS won't suffer this problem. >> You want your AFS client close to the >> user because that way the AFS Cache Manager will actually benefit your >> user. If the only contact with AFS is via Samba, you also have the >> problem that clients are unable to manage ACLs, check quota, create >> mount points, etc. since SMB/CIFS does not support those operations. > > The vast majority of users don't care about any of these operations. > They just want to save data and retrieve it. Or maybe it's save it > and forget about it. While installing the client might be a good > thing, the user isn't going to buy it based on this logic. If we had > numbers to show how much faster it is, that they would buy. Then might I suggest you start running some performance tests. Jeffrey Altman
smime.p7s
Description: S/MIME Cryptographic Signature
