# Compute RTT in theoretical system

This is problem from my Networks class. It wants us to compute the round trip time of a point-to-point network and I'm just not sure which information is pertinent. The question:

NASA deployed its latest 622Mbps point-to-point laser link between Earth and a
new lunar base. The distance from the Moon to Earth is approximately 385,000 km, and data travels over the link at the speed of light 3 x 10^8 m/s.

(a) Calculate the minimum RTT for the link

I'm not sure whether I should be using the 622 Mbps or the distance and speed of light. My initial thought was using distance and speed of light. Meaning, the minimum would be (385,000*1000)m /(3x10^8 m/s)*2 = 2.56666 seconds.

But if you use the 622 Mbps you can compute, for the minimum it would be quickest if you only sent one bit, and the other host immediately responded with one bit meaning that the RTT would be 2*(1/(622*1,000,000)) = 3.215 e-9 seconds.

To me it seems like it should be the former, but if the link is 622 Mbps why would it take longer?

EDIT: to clarify don't tell me which answer is correct, as it is homework. I would like to understand the concepts involved with why RTT is unrelated to Mbps.

• Welcome to CS.SE! Doing your exercise for you isn't very interesting, and not very useful to others in the future. Can you edit your post to ask about a specific conceptual issue you're uncertain about? As a rule of thumb, a good conceptual question should be useful even to someone who isn't looking at the problem you happen to be working on. – D.W. Sep 9 '16 at 20:03
• @D.W. I had thought it was clear I was asking about concepts, but I have added an addendum to make that explicit. – user2386276 Sep 11 '16 at 0:09
• Thanks for your comment. Honestly, "here's my exercise problem, tell me about whatever concepts I might be stuck on" isn't quite the sort of question we are ideally looking for, because it forces us to guess what concepts you need help with. It's better if you can identify what is the specific conceptual issue you are unclear on. Usually a good conceptual question shouldn't need to present any of the numbers or specifics from your exercise; it should be able to be framed in a generic way. – D.W. Sep 11 '16 at 1:39

Why would it take longer than a few nanoseconds to get a message to the Moon and back? Because physics. Nothing can go faster than the speed of light.

Instead of packets on a network, think of suitcases on the baggage reclaim belt at an airport. The number of suitcases the airport workers put on the belt each second is the data rate; the length of time it takes a suitcase to go all the way around the belt is the round-trip time.

Suppose the airport employs more staff to throw luggage around. They can now put, say, twice as many suitcases onto the belt every second, but the belt doesn't move any faster so each individual bag still takes the same amount of time to go around. They've increased the data rate but not the round-trip time. Because they're putting bags onto the belt more quickly, but the belt itself isn't moving any faster, there will be more bags in transit on the belt at any time.

Suppose that, instead of hiring more staff, they make the belt move twice as fast. Now, you don't get any more bags per second but each bag only takes half the time to go around. They've decreased the round-trip time but not the data rate. Because bags now get delivered to waiting passengers faster, but they only appear at the original rate, there will be fewer bags in transit on the belt at any time.

So, you see that data rate and round trip time are essentially independent. The data rate is how many bits you can send each second, and the trip time is how long each individual bit takes to reach its destination. You should now be able to figure out what the round trip time is, and also answer more complicated question like, "How long does it take to get a gigabyte of data to the moon on this link, and receive an acknowledgment that the final packet arrived?"

• I sort of see your point, and I definitely understand not being able to move anything faster than the speed of light but I still have some issues with this. I had thought Mbps was a property of the network/connection? But if you are saying its more like introducing information to the network then its a property of the system? That is a better processor would increase the Mbps? And the only way to get data from one place to another faster would be to shorten the distance or use a medium with better propagation speed? – user2386276 Sep 11 '16 at 0:05
• Additionally our book explains to think about bandwidth in terms of how long it takes to transmit one bit of data. To me that suggests that RTT IS dependent on the bandwidth. And that if I have a 622 Mbps link I can get a bit to the moon and back in 3.215 e-9 seconds even though that is faster than the speed of light. (Which actually suggest that a 622 Mbps link can't exist) – user2386276 Sep 11 '16 at 0:13
• @user2386276 The time taken to send more than one bit does indeed depend on the bandwidth, because that it what controls how often you can push a bit onto the wire. But, for a single bit, it depends only on how long it takes the signal to reach the other end. (Sorry for the very slow reply!) – David Richerby Sep 18 '16 at 14:48