Re: A Customer's Request For Open Source Software

[aek@vib.usr.pu.ru (Alexander Kopilovitch)]

Robert I. Eachus wrote:

> It is a resource issue.  It would be nice to have a son of Multics
> implementation in Ada, but the cost of doing that for any real platform
> is prohibitive.  I think the last time I seriously looked, Multics ran
> to over 20 million lines of (almost all PL/I) source code.  That does
> include the compilers, mail system, emacs, relational data base, etc.,
> but it gives you some idea how large the Multics environment was.

Well, this is understandable. But I guess that there may be non-conventional
ways to get something like that. One way, which I imagine, largely avoids that
"prohibitive cost" (although it still may be considered as a sort of a plot
for an episode of the serial "Wild Investor" -:) : take existing open source
OS - FreeBSD is, perhaps, the best candidate - and start to replace its layers
from the ground, that is, from the kernel, following the Multics design, and
making necessary bridges (between lower Multics-layers and upper FreeBSD
layers) at each stage, thus maintaining functionality but not too worrying
about effectiveness.

> On the other hand most RDBMSs have options for bypassing the file system
> and managing disk storage directly.  With a large amount of NVRAM, their
> is no need for the file system at all.  So my feeling is that the way to
> get something to market at the right time is to concentrate on products
> that can take full advantage of large (and non-volitile) addressable
> memories.

Well, I have vague only feeling about current state of RDBMSs, and it is not
clear for me whether people in that area will see that as major improvement
(for what I constantly hear, their main practical problem regarding efficiency
is just right index structure). But I have a friend who is an active professional
in this area, both at academic side and at practical side, and I'm going to ask
for his opinion on this matter (but it may take about 2 weeks, because just
now he is busy at a DBMS conference in Germany -;) ... if his opinion will be
positive then, perhaps, some of his Ph.D. students will become interested in
such a project -:) .



Alexander Kopilovitch                      aek@vib.usr.pu.ru
Saint-Petersburg
Russia