[OpenAFS] Ammunition

[Phil.Moore@morganstanley.com]

>>>>> "Derrick" == Derrick J Brashear <Derrick> writes:

Derrick> I don't know if it's been clarified recently but at LISA in
Derrick> 1995 Phil Moore told of Morgan Stanley's worldwide multicell
Derrick> (40ish cells) environment.

The LISA 95 paper was the "here's what we're trying to do" paper.

Since then, we've done it.  Morgan Stanley's entire distributed system
infrastructure depends critically on AFS as the primary distributed
filesystem for application deployment.

You want an application in production here?  You have to deploy it in
AFS.  We primarily leverage the readonly replicated volume
functionality of AFS, but we also have a global readwrite filesystem
implemented as well.

We've had performance and stability issues with readwrite data,
however, and strategically we're migrating most readwrite data into
NFS/CIFS.  Yeah, this comes at a huge loss in manageability and
security, but life's a bunch of trade offs...  The bottom line is that
AFS is suboptimal for readwrite data, and with the availability of
MP-fast NFSv3 servers and clients, AFS just can't compete when raw
performance is an issue.

Unfortunately, our Corporate Communications office will NOT let
OpenAFS, or anyone else, use our name and our IT infrastructure as a
case study, and that's why you don't see anything about us on the
testimonials.  I can get away with emails like this, though ;-)

Also, be careful, since using our success with AFS as an example might
very well be comparing apples to oranges when considering your own
environment.  We've heavily automated the management of AFS, and I
beleive that we stress the produce in ways that very few others do.