Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

If I've implemented a reliable UDP file transfer protocol and I have a file that deliberately drops a percentage of packets when I transmit, why would it be more evident that transmission time increases as the packet loss percentage increases going from the client to server as opposed from the server to the client? Is this something that can be explained as a result of the protocol?

Here are my numbers from two separate experiments. I kept the max packet size to 500 Bytes and the opposite direction packet loss to 5% with a 1 Megabyte file:

Server to Client loss Percentage varied:

1 MB file, 500 b segments, client to server loss 5%

1% : 17253 ms

3% : 3388 ms

5% : 7252 ms

10% : 6229 ms

11% : 12346 ms

13% : 11282 ms

15% : 9252 ms

20% : 11266 ms

Client to Server loss percentage varied

1 MB file, 500 b segments, server to client loss 5%

1%: 4227 ms

3%: 4334 ms

5%: 3308 ms

10%: 31350 ms

11%: 36398 ms

13%: 48436 ms

15%: 65475 ms

20%: 120515 ms

You can clearly see an exponential increase in the client to server group

share|improve this question
Which way was the file going? – David Schwartz Nov 21 '12 at 22:58
From the client to the server, back to the client (my outgoing server was – mighty_squash Nov 21 '12 at 23:10
So the client says the entire 1MB to the server and the server send 1MB back? Or do they overlap? – David Schwartz Nov 21 '12 at 23:14
My implementation is iterative, so I (the client) send to the file to (me) through the server, so essentially it boomerangs the file back to me. – mighty_squash Nov 21 '12 at 23:17
That sounds sufficiently symmetric that there's no reason one direction's drop should behave different than the other's. But it's hard to know without understanding the implementation details of the protocol. – David Schwartz Nov 21 '12 at 23:18

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.