Server Fault is a question and answer site for system and network administrators. 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

I am trying to reproduce a network-related performance problem from a remote site to a cloud-hosted system I'm responsible for. They get awful performance from their location, I get fine performance from mine. I have a hardware WAN emulator (Apposite Linktropy Mini2 in this case) I can use to try to simulate their network to reproduce the problem. But first, I need to know the performance characteristics of the network between them and the server.

What I need is a concise way that I could have an average user run from their PC or whatever that could sum up bandwidth, latency, packet loss, and whatever else I can get. Ideally they'd just put in a DNS name and generate some numbers they can feed me - I can't have them do all kinds of techie work, make them find a Linux box, etc.

What would a good tool be that would characterize the network between two points?

share|improve this question
To clarify my need - I need something a random human at a customer site can run and pass me their numbers, not advice on various Linux commands to run. – Ernest Mueller May 20 '11 at 3:14
up vote 3 down vote accepted

Do they have other networking issues? e.g. dropped broadband, slow performance in general etc?

You could try several runs of to the nearest location to the cloud hosted system. several runs.

Then set up ping plotter to ping the the cloud based system which will give you an idea of packet loss etc.

It's entirely possible they have other issues or a router between them and the remote system is faulty or badly set up. It may even be their own router.

share|improve this answer
It may, but the problem is remotely diagnosing it. They are in Bangalore though, so I'm banking on normal "network stretching round the world" issues, I just want to be able to emulate it. – Ernest Mueller May 20 '11 at 1:14
Also, I'm not their network admin - I want to emulate their network to see what's up with a Web service application, figure out why it's misbehaving for them. My scope is limited to that... – Ernest Mueller May 20 '11 at 13:32
Out of a desire for ease of use I went the route of having them run some and runs to the POP closest to the cloud data centers I was interested in, though I may try these other methods as well at some point! Thanks all! – Ernest Mueller May 26 '11 at 19:30
Did you say Bangalore? I wonder if it's a cabling issue? – Matt May 27 '11 at 1:37

iperf is free/open source, runs on Windows and *NIX, supports TCP and UDP to test bandwidth, latency, jitter, packet loss, etc.

share|improve this answer
I downloaded this (well, the Windows binary I found from…) but it seems to be a bit too complex for remote troubleshooting, and I'd have to run the server all the time - hoping for something more "Hey whenever you want, since you're in a different time zone and all, run this and send me the numbers for me to plug in". – Ernest Mueller May 20 '11 at 3:27
Doesn't get much easier than; that's what I'd tell them to use. – gravyface May 20 '11 at 3:36
Yeah, a pingtest/speedtest combo would get numbers to somewhere, I was just hoping to get it actually to our site. Their coverage in e.g. India is sketchy. – Ernest Mueller May 20 '11 at 13:29
if you want point-to-point numbers, well, you need to use a point-to-point tool. :) You could easily wrap up iperf along with a batch file; I've done that for laptop-to-laptop wireless testing with a non-technical helper; was click/run/done. – gravyface May 20 '11 at 15:43

It sounds like Qcheck might be the tool for the job.

share|improve this answer
Hmmm, I downloaded this but you have to install custom endpoints on every box you want to measure to, which seems unwieldy and is going to cause no end of firewall problems. – Ernest Mueller May 20 '11 at 3:05

Try mtr , it is great for this kind of thing. nmap can help with scanning the network to understand it better.

You want to analyze the paths being taken in a traceroute , you might find there is a dodgy provider in the middle somewhere dropping lots of packets or a sub-optimal path is being taken.

share|improve this answer
Got WinMTR, that's pretty good, simple enough for someone remote to run, no custom endpoints, gets packet loss and latency. Does need ICMP to work, which will be a problem in some circumstances but works for me at the moment. Now I just need throughput. – Ernest Mueller May 20 '11 at 3:24

You could use bing. It has restrictions - it uses ICMP and cannot work end-to-end if the client uses a NAT router, but you should be able to get a bandwidth estimate. If the performance is not constantly bad, you might need to use some kind of logging over an extended period of time.

If you suspect that latency or packet loss is the reason for bad performance, you might just as well inspect the TCP control flow after a packet capture - if you see slow starts or retransmissions from your side, you know for sure.

share|improve this answer

Your Answer


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

Not the answer you're looking for? Browse other questions tagged or ask your own question.