CS439: Principles of Computer Systems

Discussion Section 9 Problem Set

Due in Section on Friday, April 11, 2025

The problem set must be completed before section and brought to section. To ensure that your problem set is turned in correctly and that you receive credit for discussion section, you must follow these guidelines exactly.

Include the following header at the top of your file:

Discussion Section Problem Set #:
Student Name:
EID:

You may either type your answers or write them by hand, but you must have a copy with you in discussion section.

In discussion section, your TA will check that you made a good effort on the problem set, you are participating in the discussions in section, and you are ready and able to answer when called upon. If all of those are true, you will receive credit for discussion section that day. Note that a "good effort" on the problem set includes that all answers are in words unique to you and not copied from anywhere else, including the slide deck.

Using the traceroute command, trace the hops a packet takes from one of the UTCS machines to a website in Europe. Inspect the trace. What do you notice? Do you see any machines you recognize? ISPs? Cities? Please be sure to tell us what website you used.
Perform the trace again. How does the second time affect the output?
Describe the steps necessary to prepare a server to accept connections. How does it accept a connection?
Which is the best (fastest) network on which to implement a remote-memory read that sends a 100 byte packet from machine A to machine B and then sends a 8000 byte packet from machine B to machine A?
1. A network with 200 microsecond processing overhead, 10 Mbyte/s bandwidth, 20 microsecond latency
2. A network with 20 microsecond processing overhead, 10 Mbyte/s bandwidth, 200 microsecond latency
3. A network with 20 microsecond processing overhead, 1 Mbyte/s bandwidth, 2 microsecond latency
4. A network with 2 microsecond processing overhead, 1 Mbyte/s bandwidth, 20 microsecond latency
How does a journaling file system maintain file system consistency even in the event of a crash? Name an example of a journaling file system.
Your friend is leading the networking department at PixelForge Systems. Deadlines are coming up, so they decide to not implement congestion control in their implementation of TCP. TCP provides reliability, they think, and so what is the worst that could happen? Given that their implementation is widely adopted, provide them with an answer to their question: What is the worst that could happen?