Wget Question

[delegate]

We are downloading files using wget. At the end we want to calculate throughput for each file. So below is the output given by wget.
ftp://xxx.xxx.xxx.xxx/15MB--20:56:52
Connecting to xxx.xxx.xxx.xxx:xx... connected.
Logging in as xxxxx ... Logged in!
==> SYST ... done. ==> PWD ... done.
==> TYPE I ... done. ==> CWD not needed.
==> PASV ... done. ==> RETR 15MB ... done.
Length: 15,769,600 (15M)

100%[======================================>] 15,769,600 10.91M/s

20:56:53 (10.90 MB/s)- `15MB' saved [15769600]

real 0m1.486s
user 0m0.044s
sys 0m0.188s


what is 10.90 MB/s----what can we call it----can we call it throughput---what term best describes 10.90 MB/s---when we calculate throughput say eg 15 MB(file size)/time(0m1.486s) we are getting a figure different from 10.90MB/s

So what can we call 10.90MB/s---coz we want also to graph it.

If 10.90 MB/s is throughput how can we separate it from---throughput we are calculating using the formula(15 MB(file size)/time(0m1.486s) )

Your help is greatly appreciated we are newbies

[djgrandmarquis]

I think "throughput" is an accurate term.

Keep in mind that there are other things going on that may introduce latency and influence the real time of the download. There may have been packet loss or data corruption that triggered retransmission of some packets. Other processes on your machine may have been consuming CPU resources, etc.

It's difficult to know exactly what the FTP client is calculating. You can go poking around the specification and implementation of the FTP protocol, but the easiest thing to do is just use the FTP client's results. Do several trials, average them, and you should have reasonable, comparable results.