CS261 Fall 2013: Quiz #1
This exam is closed notes, closed papers, closed everything.
Please use whatever word processing software that you'd like
to write your answers, but please send me either plain text
Answers should be emailed to: email@example.com.
Here are the papers we've read:
- Baumann: The Multikernel: A New OS Architecture for Scalable Multicore Systems
- Bugnion: Disco: running commodity operating systems on scalable multiprocessors
- Corbató: An Experimental Time-Sharing System
- Creasy: The Origin of the VM/370 Time-Sharing System
- Daley: Virtual Memory, Processes, and Sharing in MULTICS
- Dijkstra: The Structure of the "THE" Multiprogramming System
- Gamsa: Tornado: Maximizing Locality and Concurrency in a Shared Memory Multiprocessor
- Kilburn: One Level Storage System
- Leidtke: On micro-kernel Construction
- Rashid: From RIG to Accent to Mach: The Evolution of a Network Operating System
- OS Intellectual Heritage
The UNIX time-sharing system was both a departure from and a natural successor
to many systems that came before it.
Drawing on all the papers we've read, identify all the systems that might be
considered intellectual ancestors of UNIX.
For each one, include a sentence or two explaining why you listed it.
Now, identify those systems that might be considered intellectual descendents of UNIX
and provide a sentence or two explanation of why you think they might be considered
Compare and contrast a multikernel (Barrelfish), clustered object (Tornado),
and a hypervisor-based approach to scaling on multiple processing elements
Discuss the key abstractions used in each, how those abstractions are
like/unlike those in the other systems, and whether there are fundamental
differences or merely implementation differences among them.
- Walk a Mile in another's shoes
Imagine you are Jochen Leidtke. Write a one paragraph critique for
each of the Dijkstra and Rashid papers (don't forget that the Rashid paper included
three systems and their evolution -- it's not just a paper about Mach).