On Fri, 15 Oct 1999, Miles Nordin wrote:

> Evangelism aside, pulling this up to -current, playing with it, optimizing
> it, u.s.w., might be an interesting and fun project for someone interested
> in theoretical CS, as well as extremely useful to the hopefully-increasing
> numbers of people who use NetBSD for their routing (what with the
> excellent built-in IPv6 support and all).
> 
> At the very least, it's ``Research Done Using NetBSD,'' and TNF might want
> to brag about it on the web site.  anyway, have a look:
> 

I know it's not as sexy as network research, but at NASA Ames we've poured
a ton of effort in making changes the vnode/vfs stuff as part of our
Heirarchical Storage Management, HSM, system. NetBSD is the first system
that it will work on and is where the research will take place. Someday,
once it's been proven, it could certainly be ported to Solaris, and other
BSD's, but we have chosen NetBSD for now because of the super-clean nature
of the kernel. 

Hopefully, we'll have a paper ready for the next IEEE storage conference.
Bill Studenmund is working on other papers as well on his vnode changes
and our DMFS, data migration file system, layer. All of this will
prominently highlight netbsd. A lot of other work is being done as
well by Matt Jacob and Jason Thorpe to make all of the other needed
infrastrure gel, along with a lot of application code that is being
written by people that you likely aren't aquainted with. Soon, people can
get beta versions. [people can email ne at waddell@nas.nasa.gov if they
are interested in more information. there is still some legaleze
that needs to get worked out] I expect that some computer centers will be
very interested in this. If so, they will have to either install a
netbsd box in a very prominent role, or have to port it to some other os
that supports stackable filesystem layers. [e.g. not linux ]

At least that's the plan. It is for the government, so some random bad
thing can always happen...

Harry Waddell