.. (לתיקייה המכילה) | ||
Should I include q in the ideal "1-p" utilization value for the graphs? | |
No, the p is exact parameter given. |
If a message was sent successfully but the ack was lost, does it contribute to the utilization? | |
It does, but consider how it affects the algorithm. If we're sending a message for the second time, this isn't a good use of the bandwidth. You may want to simplify things for the simulation for handling lost acks in this regard. |
What should be the window size for the GBN algorithm? | |
It's up to you. The choice of N should lead to the best possible utilization - just as we have done in class - while considering the properties of the simulation. |
What is the definition of "ניצולת" for the exercise? | |
The same as in class. We are interested in knowing the number the average number of messages sent successfully in the duration it takes to send a single message. You'll have to figure out how the simulations estimate this amount. |
What is a reasonable run time for the simulation? | |
The simulation should complete in 30 minutes, which is much more than needed. |
What should be the order of the results of all the different simulations in the file? | |
It doesn't matter. |
What are the intended combinations for the receiver and sender buffer sizes in SR? | |
All the combinations of integers from 1 to 15, separately for the sender and the receiver. |
In the final section, we are requested to plot the curves for Tp=1.5, yet GBN and S&W aren't tested on this value. What should be done? | |
Run the simulations again for this value, just for the graph (or add it to the output files as well, it doesn't matter). |