September 25, 2003

Notes: Definitions

Paul Walker contributes a definition of a distributed computational system:

A distributed system is one in which the failure of a computer you did not even know existed can render your own computer unusable.

Posted by DeLong at September 25, 2003 05:03 PM | TrackBack

Comments

You're finding ot about this *now*?! You have led a most fortunate life. How did you do it?

That definition is at least 25 years old, and probably older.

Posted by: Jonathan Goldberg on September 25, 2003 06:43 PM

By always using non-distributed computing systems?

Posted by: Brad DeLong on September 25, 2003 07:26 PM

Of course, the counterexample is Google, where a lot of low quality, low reliability components, by being distributed together, creates a high reliability system.

Posted by: Nicholas Weaver on September 25, 2003 07:31 PM

Leslie Lamport, a genuine hero.

Posted by: Russell L. Carter on September 25, 2003 08:29 PM

that definition is missing a word, it should read:

"can not render"

true distributed systems dont care about singularities. any system that does is an impostor.

Posted by: sampo on September 25, 2003 11:09 PM

that definition is missing a word, it should read:

"can not render"

true distributed systems dont care about singularities. any system that does is an impostor.

Posted by: sampo on September 26, 2003 12:57 AM
Post a comment