CIFS vs. NFS and other filesystems (was Client for Samba Netw orks)

Mayers, Philip J p.mayers at ic.ac.uk
Tue Dec 18 10:08:06 GMT 2001


The FileSystem driver development kit is extremely expensive. Not to mention
that the integration with the NT kernel is problematical at best - IIRC, the
SMB redirector uses undocumented entry points. And, the final killer, the 9x
kernel is extremely difficult to integrate such stuff with - and the 9x
kernel *is* a big market.

Regards,
Phil

+------------------------------------------+
| Phil Mayers                              |
| Network & Infrastructure Group           |
| Information & Communication Technologies |
| Imperial College                         |
+------------------------------------------+ 


-----Original Message-----
From: Steve Langasek [mailto:vorlon at netexpress.net]
Sent: 18 December 2001 17:23
To: samba-technical at lists.samba.org
Subject: Re: CIFS vs. NFS and other filesystems (was Client for Samba
Networks)


On Tue, Dec 18, 2001 at 10:48:50AM -0600, Steven French wrote:
> >Subject: Re: Client for Samba Networks

> >On Mon, Dec 17, 2001 at 01:31:10PM -0600, Christopher R. Hertel wrote:

> >> SMB is icky, NetWare is even more closed, and NFS is okay, but somewhat
> >> outdated.  Coda has lots of problems, but one thing it has in its favor
> >> is disconnected operation.

> > Tim Potter wrote:
> >Have a look at nfs v4.  It's very nice both compared to SMB and previous
> >versions of nfs.

> Sounds like we are overdue for a high level comparison of network
> filesystems.   Here are some
> initial observations:

> CIFS:
>      Strengths:
>      a) Huge installed client base (not just Windows),
>      b) good, open source server implementation available (Samba!),
>      c) token management (oplock) and referral ("dfs") semantics are a
good
> compromise between usefulness and simplicity
>      d) the key part of the filesystem protocol (mostly) documented,
>      rich file open semantics map well to Windows and related OSs,
>      e) kerberos security integration and RPC integration
>      f) broader in scope (print, ACL, browsing etc.) than other filesystem
> protocols
>      g) optional PDU signing above the RPC allowing maximal flexibility
>      h) Unicode

The use of UCS-2 is as much a strength as it is a weakness; Unicode does
not stop at the 16-bit boundary, and the lack of support for the
additional codepoints is of some concern to users of modern non-Western
languages.  Of course, UCS-2 is also at the heart of Windows Unicode
support, so any network fs that has to handle Windows systems will at
lesat suffer from this flaw on the client end.

> NFSv3:

>      f) relatively weak open source server implementation (at least
> compared to Samba and AFS) has scalability problems

If I'm not mistaken, much of this is due to the fact that every Open
Source platform has their own implementation (or two).  Samba and AFS
reflect much better pooling of coding resources.

Steve Langasek
postmodern programmer




More information about the samba-technical mailing list