The difference between data rate and latency

latencyterminology

Both attempt to measure the speed of a network. If data rate is the thing that is commonly measure in Mbps and latency is the thing that is measured in milliseconds, what are their similarities and differences?

I've tried searching google but the explanations I get are inconsistent. After searching "network latency" the first result is wikipedia but this definition pertains more to structural engineering than networks. The second result is about.com and it states

The term latency refers to any of several kinds of delays typically
incurred in processing of network data.

So does this mean latency is not a well-defined term and can mean whatever a person wants it to?

For example, when using ping what does the latency actually measure (and out of curiosity why doesn't ping tell the data rate, in Mbps)?

What is the perceived difference between latency and data rate, is it that latency only matters until the download (or uploaded) first part has been received?

Best Answer

Latency is the amount of time it takes a packet of data to leave your computer and receive a response back from the end point. That is why this is measure in time. This is key for time sensitive applications like VoIP and video conferencing.

Bandwidth/data rate is the amount of data (bits) you can upload or download in a given time (seconds). This is key for downloads and uploads (i.e. video streaming for Hulu, Netflix, etc). It is also important in video conferencing as you're sending/receiving a lot of video (bandwidth) and the audio portion needs to be in near real time (latency).

Ping is sending packets using ICMP and measuring the time it takes until you receive an echo-reply from the end point. It is not sending a lot of data and would not give any accurate information regarding bandwidth.

Something like iperf may be used to measure bandwidth by sending a lot of data to the end point which also would be running perf.

You can have high latency (slow round trip time for packets) but still have a lot of bandwidth. Satellite service is one example.

Related Solutions

TCP – Difference Between RWND and RWIN

From the diffs of draft-ietf-ippm-tcp-throughput-tm:

This diff seems to show all the RWIN abberviations being replaced with RWND, below is an example.

    It is important to clarify the interaction between the sender's Send
-   Socket Buffer and the receiver's advertised TCP RWIN Size.  TCP test
-   programs such as iperf, ttcp, etc. allow the sender to control the
+   Socket Buffer and the receiver's advertised TCP RWND Size.  TCP test
+   programs such as iperf, ttcp, etc. allows the sender to control the

The latest version of this document, RFC 6349, doesn't have RWIN anywhere in it. Looks like they mean the same thing to me.

Latency – Difference Between Public and Internal IP Addresses

Technically yes there will be a small difference and notability will depend on your devices/configuration. This is because of the different paths the packets have to take, but like i said, depends on how your setup is designed - there are plenty of variables.

1) If you are on the LAN then your path to the webserver via its private IP is merely just switched.

2) If you are on the LAN and you try and access the webserver via its public IP then the traffic has to go out through your LAN gateway (which I am assuming is your router with a public ip address on the other side) , get natted out, get natted inbound and forwarded to the private IP address of the server and then return the traffic.

So you can see there will be marginally more resources used than #1

Best Answer

Related Solutions

TCP – Difference Between RWND and RWIN

Latency – Difference Between Public and Internal IP Addresses

Related Topic