Amoeba (operating system)

The current title ("A way to bring back Moore's law") is somewhat misleading. You could say that Amoeba was a way to bring back Moore's law at the time, but then again, in the 1990s it wasn't exactly threatened. Nowadays the techniques used in Amoeba have spread to your "regular" Linux computing clusters. I've done some work in distributed computing and I can say that sophisticated load balancing and, to a lesser degree, process migration are known and used in the field. Lots of solutions for both for the commonly used MPI family of frameworks have been published [1].

It all comes down to Gustafson's Law [2].

[1] E.g., http://capinfo.e.ac.upc.edu/PDFs/dir25/file003041.pdf, http://www.mosix.org/pub/Process_Migration_for_OpenMPI.pdf, etc.

[2] https://en.wikipedia.org/wiki/Gustafson%27s_Law

Re: Amoeba (operating system)

#5

Post by zaptheimpaler » Fri, Mar 08, 2013, 3:58 AM UTC

pjscott wrote: post #2

There's only so much parallelism you can exploit, especially with network lags adding overhead. http://en.wikipedia.org/wiki/Amdahl%27s_law

Thats a pretty trivial bound though. It basically deals with some portion "p" of the program that can be parallelized and tells you how that would affect performance given that the rest (1-p) can't be parallelized. "p" would probably be a huge chunk of the program in most cases, say a DB lookup if that is the biggest bottleneck.

Re: Amoeba (operating system)

#6

Post by jacques_chester » Fri, Mar 08, 2013, 4:18 AM UTC

My first instinct was to bust out a link to the "Network Fallacies" paper and a snarky remark, but it occurred to me that it's due to projects like Amoeba, Sprite and Plan 9 that we even really understand those fallacies.

Re: Amoeba (operating system)

#7

Post by jacques_chester » Fri, Mar 08, 2013, 4:32 AM UTC

DennisP wrote: post #3

> The Python programming language was originally developed for this platform That just struck me as a really nifty bit of trivia.

Tcl was developed for another distributed OS project, Sprite.

And of course C was developed for Unix.

What is it about operating systems and programming languages? :D

Re: Amoeba (operating system)

#8

Post by kunai » Fri, Mar 08, 2013, 4:47 AM UTC

jacques_chester wrote: post #7

DennisP wrote: post #3

> The Python programming language was originally developed for this platform That just struck me as a really nifty bit of trivia.

Tcl was developed for another distributed OS project, Sprite. And of course C was developed for Unix. What is it about operating systems and programming languages? :D

AFAIR Java wasn't really developed for any specific OS.

Re: Amoeba (operating system)

#9

Post by malandrew » Fri, Mar 08, 2013, 4:53 AM UTC

jacques_chester wrote: post #7

DennisP wrote: post #3

> The Python programming language was originally developed for this platform That just struck me as a really nifty bit of trivia.

Tcl was developed for another distributed OS project, Sprite. And of course C was developed for Unix. What is it about operating systems and programming languages? :D

Interestingly you can say the same about Rust and Servo. After all the browser is essentially becoming an operating system with an entirely different trust model.

Re: Amoeba (operating system)

#10

Post by rozim » Fri, Mar 08, 2013, 4:55 AM UTC

kunai wrote: post #8

jacques_chester wrote: post #7

Earlier quoted context omitted.

Tcl was developed for another distributed OS project, Sprite. And of course C was developed for Unix. What is it about operating systems and programming languages? :D

AFAIR Java wasn't really developed for any specific OS.

I thought it was developed for settop boxes - and called 'Oak' back then..