Project Athena ( was Re: Non Destructive Version of rm)

[The Grand Master]

In article <{#_**}@ads.com> henry at ADS.COM (Henry Mensch) writes:
}background:  i'm a former staff member of MIT's Project Athena and of
}the Purdue University Computing Center, and am familiar with the styles
}of computing available in both places.  i represent only my personal
}opinion with this message, and do not represent the opinion of Project
}Athena or the Institute.
}
}asg at sage.cc.purdue.edu (The Grand Master) wrote: 
}
}everyone out there who uses sun's automounter is doing exactly the same
}thing (they are, of course, limited to the one networked file system
}that is supported by the automounter).  it works quite well.  i suggest
}you try it.

As you point out later, I believe I have.
}
}->You do not however have any more power (in fact quite a bit less) than systems
}->which now occupy about the same volume as my desk.
}
}consider this measure:  divide the number of MIPS that your
}timesharing processor provides by the number of users  sharing that
}processor.  apply the same measure to the workstation user (remember
}that only one user uses a workstation at any one time).

If I have a Sequent witht the same number of processors in it as
the number of users (40 max I believe) I would expect the numbers
to come out pretty even - a little higher on the side of the 
Sequent, since it is running only 1 init for 40 people (and one
inetd, and one nfsd etc) while the workstation idea has to run
40 of these for 40 people. What you also forget to mention here is
what happens when only 3 people are logged into a 40 processor system.
The sep workstation method takes no advantage (or very little) in
only 3 of 40 workstations being used, but if only 3 people log into
my 40 processor sequent, it will run much faster than a workstation.
}
}->You can have a large scale distributed environment without allowing
}->people to mount their own directories. We have such at Purdue with the
}->few centralized systems (yes, you can all sing along) with Xwindows
}->terminals.
}
}this is not a distributed computing environment.  replacing character
}terminals with x terminals (did you know that the x window system came
}from project athena and is an athena network service?) does not make a
}distributed computing environment.
}
Well, then i guess I do not have the correct definition of distributed
environment. The effect is basically the same though as I see it.
}->We also have it here at GE where each person who has a workstation
}->can still log into anny workstation and be able to access his disk without
}->having to do mounting all over the place. If I want to get to a directory
}->/tmp on the system a294 I do cd //a294/tmp - no problem.
}
}and how do you think //a294/tmp gets there?  magic?  maybe the
}workstation *gasp* mounts a filesystem there for your use when you
}make such a request.  maybe they're using an automounter?

May be, but I am still not allowed to decide where to mount file system's
(that is preordained) and I do not have the root password.
}
}->It is ALREADY WORKING! 
}
}nobody said it wasn't working.  you didn't address jik's point that
}this solution did not scale to hundreds of servers and thousands of
}hosts.  i do believe you only have a handful of sequents at purdue ...

True, but that does not mean it will not scale to a larger size. 
As long as entombd knows what filesystem it has mounted and where
it is mounted from, then the rpc instructions will tell the
server what to d o about entombing
}
}->Oh, I like your setup even better now. Give all the users root! 
}
}you obviously don't understand that (with single user hosts) any user
}can become root with little/no effort.  project athena's architecture
}gets over this by making root (on a workstation) a commodity of
}limited value.  our users don't get root privileges on servers (where
}data are kept and network services originate from).
This is not neccessarily true. Please explain to me why it is 
impossible to keep a single user system secure. in any event, 
using Xterminals with some centralized system, it IS possible
to keep people from becoming root.
}
}bruce, you might do well to have a chat with the manager of UNIX
}systems at the computing center ... i know he has some clues about
}
Here we go again. Did you learn that tactic from Jon? Just because
I disagree with you does not mean I am incapable of understanding.
I know full well the merits of Athena's setup. But the biggest
problems I have with Athena are
1) If I am using Xwindows, and I do something CPU-intensive, I cannot
  even get my pointer to move around at times, much less iconify
move or size windows. With Xterminals, since the graphics are handled
by the terminals processor, the function of your windowing environment
is (somewhat) independant of the load on the man system.
2) Athena's setup does not take advantage of unused resources when
 only a few people are logged in. My machine is just as slow when
I do a CPU intensive job whether all 1000 workstations are in
use, or if only 100 are in use. With the method that I advocate however,
My job can take advantage of those unused resources. 
}--
}# Henry Mensch / Advanced Decision Systems / <henry at ads.com>

---------
                                   ###             ##
Courtesy of Bruce Varney           ###               #
aka -> The Grand Master                               #
asg at sage.cc.purdue.edu             ###    #####       #
PUCC                               ###                #
;-)                                 #                #
;'>                                #               ##