12

I have a web server in Ireland (Amazon AWS). This server appears fast from Germany (orange line) but slow from the USA (black line). The HTTP request used for the test is the same.

enter image description here

I think this is normal. The distance between Ireland and the USA is larger than Germany to Ireland, but the difference seems too high.

Are there other possible reasons, apart from the distance to the server?

  • 7
    What is your graph supposed to be showing? round trip time? total request time? something else? – Peter Green Jul 1 '16 at 19:57
  • @PeterGreen http request time – ar099968 Jul 8 '16 at 7:08
24

Assuming the graph is http request time it seems fairly reasonable to me.

A http request (in the absense of keepalive, fastopen etc) normally requires at least two round trips.

  • Client sends syn
  • Sever receives syn and sends syn-ack
  • Client receives syn-ack and sends ack and request.
  • Server sends response.

The speed of light in fiber is about 2*10^8 meters per second. According to google the distance from "ireland to the USA" is 6,629 km * which would translate to a round trip time of about 66 ms.

But that assumes there are no delays in equipment and that the data route follows the shortest possible path. Practical round trip times are usually 100 to 150 milliseconds between a host in Europe and a host in the USA. As such a http request time of ~250ms is perfectly normal.

What is a bit more concerning are the spikes in the graph, they suggest network congestion somewhere between the server and the test client.

* obviously it depends on what point in the USA and what point in ireland but the point google picked seemed to be somewhere in the middle of the USA and the OPs graph said "us-mid".

19

Any other possible reason in addition to distance from server?

The path the packets take.

  • 6
    ...and all of the devices in between. – EEAA Jul 1 '16 at 16:42
  • 8
    ... and how congested they are – Iain Jul 1 '16 at 16:43
  • 1
    How is that relevant to the question asked... it is what it is surely? Have you considered using a tool like say mtr to examine the paths taken and the timing of each hop? – Iain Jul 1 '16 at 17:04
  • 13
    I see this kind of question all the time where I work, giving nice graphics that tend to give the impression that the user knows what they're talking about up until you ask them what traces and traffic checks they did and nothing was checked. – Julie Pelletier Jul 1 '16 at 17:34
  • 4
    @Julie yes the graphs are meaningless without knowledge and understanding. – Iain Jul 1 '16 at 18:16
9

Searching google for ping "us-mid" yields Monitis.com and their IP in Dallas, who also have a DE IP in Frankfurt. From a very well-connected server in France I have 9 ms ping RTT to the DE IP and 111 ms ping RTT to the US-MID IP. For HTTP time-to-response I'd expect double that plus the server's reaction time, for the sake of argument let's say maybe 26 and 230. That's not so different from your values, given that in Ireland you are probably a bit further network-wise from both places.

Accept those values as normal and monitor deviations from them to see it there are problems on your site or with your ISP. Unless you're running a global service that's really response-time critical, it's useless to do more. If you are, put servers around the world or (better) contract with hosters who specialize in doing that.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

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