Data rate misconception - General Question

Answered Question
Sep 28th, 2009
User Badges:

I was wondering if someone could clarify this for me?


Let's say Source host and destination host are the same distance apart for both scenarios. We'll say 100 meters.

We'll also say there are no networking problems and the same amount of intermediary networking devices between Host A and Host B.

So basically, all things physical and logical are equal for the two Scenario's except for "link speeds". Or should I say, "link capacity."


No other traffice excpet for the 1 packet being sent.


Scenario 1:

Host A sends 1 packet to host B over a 10Mbs copper ethernet link.


Scenario 2:

Host A sends 1 packet to host B over a 100Mbs copper ether net link.


1.) Isn't it true that the packet should get from Host A to Host B in the same amount of time regardless of the link speed being 10Mbs vs 100Mbs?


2.) Shouldn't this be true up until the 10Mbs link nears it's capacity of 10Mbs?


I've tried to google information about this but cant seem to find any references. Does anyone have any documentation or can point me to any refrences regarding Data-rate and Link speeds?


Much appreciated!


Correct Answer by Joseph W. Doherty about 7 years 7 months ago

Correct, as long as "amount" is measure for the same time period, and there isn't, BTW, a threshold.


"Speed" is often used rather that "bandwidth", but this is often misleading. A 100 Mbps link can "speed" a file transfer vs. using a 10 Mbps link because the transfer could take 1/10 the time; it's "faster". Yet for interactive network access, there's often little improvement using a link of more bandwidth (if distance latency is the issue, not capacity).


This explains why many who try to improve "slow" network applications find adding more bandwidth often doesn't "speed" the application.

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 4 (3 ratings)
Loading.
paolo bevilacqua Mon, 09/28/2009 - 08:14
User Badges:
  • Super Gold, 25000 points or more
  • Hall of Fame,

    Founding Member

1) no. there is a small difference due to different serialization delay.


2) no. (1) is true under all circumstances.

mortonjes Mon, 09/28/2009 - 09:33
User Badges:

Thank you for your reply. If you have a moment, please see my follow-up question...

Joseph W. Doherty Mon, 09/28/2009 - 08:25
User Badges:
  • Super Bronze, 10000 points or more

Everything else being equal, the leading edge of the frame/packet will, more or less, arrive at the same time. However, the trailing edge of the higher bandwidth will arrive sooner. For 10 vs. 100 Mbps, the tail of the 100 Mbps frame/packet will arrive in 1/10 the time. (Or to put it another way, we can transfer 10 frames/packets at 100 Mbps in the same time as 1 frame/packet at 10 Mbps.)

mortonjes Mon, 09/28/2009 - 09:30
User Badges:

Thank you for your explanation. So if I'm understandig you correctly, the "speed" at which a frame/packet traverses the link is "roughly" the same but the "amount" of data to travers the link is "more." (Up until a certain threshold of course.)


1 frame/packet @ 10 Mbps

10 frames/packets @ 100Mbps


Much appreciated.


Jesse

Correct Answer
Joseph W. Doherty Mon, 09/28/2009 - 12:11
User Badges:
  • Super Bronze, 10000 points or more

Correct, as long as "amount" is measure for the same time period, and there isn't, BTW, a threshold.


"Speed" is often used rather that "bandwidth", but this is often misleading. A 100 Mbps link can "speed" a file transfer vs. using a 10 Mbps link because the transfer could take 1/10 the time; it's "faster". Yet for interactive network access, there's often little improvement using a link of more bandwidth (if distance latency is the issue, not capacity).


This explains why many who try to improve "slow" network applications find adding more bandwidth often doesn't "speed" the application.

Actions

This Discussion