| 1 |
J. Roeleveld <joost <at> antarean.org> writes: |
| 2 |
|
| 3 |
> AFS has caching and can survive temporary disappearance of the server. |
| 4 |
|
| 5 |
Excellent for low bandwidth connections. Most DFS have mechanisms to |
| 6 |
deal with transient failures, but not as generaous on the time-scale |
| 7 |
as AFS. I believe, if I recall correctly, these hi-latency, low bandwith |
| 8 |
recovery mechanism keen design paramters, at least bake in the |
| 9 |
CMU develop cycples, for AFS? |
| 10 |
|
| 11 |
While attractive for your situation, these features might actually |
| 12 |
be detrimental to a hi_performance distributed cluster's needs for |
| 13 |
a DFS? |
| 14 |
|
| 15 |
|
| 16 |
> For me, I need to be able to provide Samba filesharing on top of that |
| 17 |
> layer on 2 different locations as I don't see the network bandwidth to |
| 18 |
> be sufficient for normal operations. (ADSL uplinks tend to be dead slow) |
| 19 |
|
| 20 |
Yea, I'm not going to be testing OpenAFS for my needs, unless I read |
| 21 |
some compelling publish data on it's applicability to high end |
| 22 |
clusters best choice as a DFS..... |
| 23 |
|
| 24 |
It's probably great for SETI etc etc. |
| 25 |
|
| 26 |
|
| 27 |
James |
Archives