Project grading standards


NETWORK WORKBENCH GRADING PROCEDURES FOR MARK PULLEN'S CLASSES

Requirements for acceptance: All projects are to be submitted using the 
with the module name as established in NW (e.g. "BitStuffing.java"). Check 
the code file to be sure it is original and well commented. Cut up to 
40% (generally this will be 1 or 2 points) for insufficient comments.
The essence of these projects is for the student to understand and
clearly convey how to make the protocols work.

Lateness: If the date/time the email was sent is later than the due 
date/time for the assignment, cut ten percent of the full possible grade 
(in most cases this will be .5 points) per week.

Specific results expected for each assignment are listed below. If the 
expected results are obtained give full credit (less any cuts above).
Here are the standards for partial credit:

 No code other than the stub submitted:                  no credit

 Program at least compiles and has adequate comments:          40%

 Program runs but output does not show expected results:       60%

 Program runs and output is partially correct, or is correct
 but code has a clear deficiency that was not tested:          80%

 Where the assignment includes questions, the questions count
 20% of the grade.

The grader may comment on particular errors, but that is not required; 
the grade is based on the code and output files submitted, so the grader 
is required only to describe how the program falls in one of the above
categories. Unusual cases that do not fit the grading standard are to 
be forwarded to the professor for evaluation.
Expected results for the various projects are:

DLC1 - bits show as stuffed in the binary frame, in the correct
        position

DLC2 - the CRC FCS in the last 16 bits of the frame (before the
        ending flag) is correct with required test input
DLC3 - the reliable DLC transfers exactly one copy of each sent "email” between host 1.1 (which is router1) and host 2.1 (which is router2), despite errors in data transmission

 

LAN1 - the Data Frames Sent & Received and Collisions Detected at the
        end of the NW output match the NW solution; also the backoff
        process messages in the output show the same pattern as the 
        NW solution; ends with the same timing as the JNW solution

TRN1 - the 'email' file transfers; the student answers the questions
        correctly

WAN2 - the packet forwarding matrix output is correct; the 'email' is
        routed to destinations properly
Last modified: Saturday, 21 January 2017, 11:11 AM