Network latencies between opposite ends of the Earth
This is just out of curiosity.
I'm currently in South-east asia, and when I pinged domains such as bbc.co.uk and google.com (in California), I get latencies of around 5ms as below:
64 bytes from 151.101.192.81: icmp_seq=0 ttl=55 time=2.940 ms
64 bytes from 151.101.192.81: icmp_seq=1 ttl=55 time=3.785 ms
64 bytes from 151.101.192.81: icmp_seq=2 ttl=55 time=6.299 ms
64 bytes from 151.101.192.81: icmp_seq=3 ttl=55 time=4.065 ms
64 bytes from 151.101.192.81: icmp_seq=4 ttl=55 time=4.231 ms
I'd expect latencies of at least 50ms instead given the geo distance.
What's going on behind the scenes?
Also is the traffic via trans-ocean fibre optics?
network latency
add a comment |
This is just out of curiosity.
I'm currently in South-east asia, and when I pinged domains such as bbc.co.uk and google.com (in California), I get latencies of around 5ms as below:
64 bytes from 151.101.192.81: icmp_seq=0 ttl=55 time=2.940 ms
64 bytes from 151.101.192.81: icmp_seq=1 ttl=55 time=3.785 ms
64 bytes from 151.101.192.81: icmp_seq=2 ttl=55 time=6.299 ms
64 bytes from 151.101.192.81: icmp_seq=3 ttl=55 time=4.065 ms
64 bytes from 151.101.192.81: icmp_seq=4 ttl=55 time=4.231 ms
I'd expect latencies of at least 50ms instead given the geo distance.
What's going on behind the scenes?
Also is the traffic via trans-ocean fibre optics?
network latency
I've run tests that don't involve CDNs and I'd estimate you're more likely to have latencies of 250 - 500ms between you and servers actually in Europe. Perhaps less, but you're right to think that global distances have effects on minimum latencies that no network topology can reduce.
– Todd Wilcox
May 14 at 16:15
Why don't you provide traceroute full output?
– Satish
May 20 at 20:31
add a comment |
This is just out of curiosity.
I'm currently in South-east asia, and when I pinged domains such as bbc.co.uk and google.com (in California), I get latencies of around 5ms as below:
64 bytes from 151.101.192.81: icmp_seq=0 ttl=55 time=2.940 ms
64 bytes from 151.101.192.81: icmp_seq=1 ttl=55 time=3.785 ms
64 bytes from 151.101.192.81: icmp_seq=2 ttl=55 time=6.299 ms
64 bytes from 151.101.192.81: icmp_seq=3 ttl=55 time=4.065 ms
64 bytes from 151.101.192.81: icmp_seq=4 ttl=55 time=4.231 ms
I'd expect latencies of at least 50ms instead given the geo distance.
What's going on behind the scenes?
Also is the traffic via trans-ocean fibre optics?
network latency
This is just out of curiosity.
I'm currently in South-east asia, and when I pinged domains such as bbc.co.uk and google.com (in California), I get latencies of around 5ms as below:
64 bytes from 151.101.192.81: icmp_seq=0 ttl=55 time=2.940 ms
64 bytes from 151.101.192.81: icmp_seq=1 ttl=55 time=3.785 ms
64 bytes from 151.101.192.81: icmp_seq=2 ttl=55 time=6.299 ms
64 bytes from 151.101.192.81: icmp_seq=3 ttl=55 time=4.065 ms
64 bytes from 151.101.192.81: icmp_seq=4 ttl=55 time=4.231 ms
I'd expect latencies of at least 50ms instead given the geo distance.
What's going on behind the scenes?
Also is the traffic via trans-ocean fibre optics?
network latency
network latency
asked May 14 at 12:51
MaveMave
16613
16613
I've run tests that don't involve CDNs and I'd estimate you're more likely to have latencies of 250 - 500ms between you and servers actually in Europe. Perhaps less, but you're right to think that global distances have effects on minimum latencies that no network topology can reduce.
– Todd Wilcox
May 14 at 16:15
Why don't you provide traceroute full output?
– Satish
May 20 at 20:31
add a comment |
I've run tests that don't involve CDNs and I'd estimate you're more likely to have latencies of 250 - 500ms between you and servers actually in Europe. Perhaps less, but you're right to think that global distances have effects on minimum latencies that no network topology can reduce.
– Todd Wilcox
May 14 at 16:15
Why don't you provide traceroute full output?
– Satish
May 20 at 20:31
I've run tests that don't involve CDNs and I'd estimate you're more likely to have latencies of 250 - 500ms between you and servers actually in Europe. Perhaps less, but you're right to think that global distances have effects on minimum latencies that no network topology can reduce.
– Todd Wilcox
May 14 at 16:15
I've run tests that don't involve CDNs and I'd estimate you're more likely to have latencies of 250 - 500ms between you and servers actually in Europe. Perhaps less, but you're right to think that global distances have effects on minimum latencies that no network topology can reduce.
– Todd Wilcox
May 14 at 16:15
Why don't you provide traceroute full output?
– Satish
May 20 at 20:31
Why don't you provide traceroute full output?
– Satish
May 20 at 20:31
add a comment |
6 Answers
6
active
oldest
votes
Google in particular uses distributed datacenters around the globe.
They announce the same IP network at various places and due to the way routing protocols work, you reach the nearest one.bbc.co.uk points to an IP address that belongs to Fastly, Inc, a content delivery network, that also has points of presence around the world, including Asia, but I don't know if they use the same technique.
(From France I have around 60ms to bbc.co.uk (151.101.192.81))
However, due to the extremely low latency you see, my bet is that you are not contacting the actual servers but this is a proxy that responds to you.
Proxy or load-balancer, which we use a lot based on which country you come from.
– user56700
May 14 at 13:06
2
It doesn't have to be a strange proxy server; it's normal to have 4-5 ms latencies to a CDN, if the ISP directly peers with that CDN and your own uplink isn't Wi-Fi or ADSL...
– grawity
May 14 at 16:31
add a comment |
A cable half-way around the globe has a minimum latency of 100 ms, 200 ms round-trip (20,000 km distance / 200,000 km/s signal speed). In reality, links aren't as the crow flies (at all) and there are additional, active components in between, adding to latency - a more realistic figure is 200 or even 300 ms one-way.
Anything with lower latency or ping-time is closer to you. As Ron's already pointed out, large service provider use globally distributed infrastructure with location-specific DNS resolution or anycasting to connect you to a nearby server.
2
As a (small sample size) test, Wellington-Madrid (very close to opposite) is about 300ms round trip pic.nperf.com/r/3190596119915611-Bc4TJSUS.png pic.nperf.com/r/3190596597699589-shrThPRn.png
– Someone Somewhere
May 15 at 8:21
add a comment |
Major domains have servers around the world, and DNS set up so that you get an IP address that's close to you.
You can avoid this by pinging a specific host. For example, the Debian (GNU/Linux) project has a list of mirrors in countries world-wide. Most of those are specific hosts that will look up to the same IP regardless of where you are. Especially university software mirrors are almost never behind a content-delivery proxy network.
For example, mirror.csclub.uwaterloo.ca
is in Waterloo, Ontario, Canada, and does respond to ping requests. I get ~37ms ping time to it right now (early morning) from Halifax, NS, Canada.
mirror.aarnet.edu.au in Australia is probably one of the farthest away from me; Australia as a whole is network-wise far from most of the rest of the world. (although South-East Asia is closer than most to it) My ping time is ~295ms.
And yes, speed-of-light delays imposed by distance around the earth's circumference are a major part of that, as well as routing delays on hops. (And remember, it's speed of light in glass fiber not vacuum. The index of refraction of the core of an optical fiber is often something like 1.3 to 1.4, so speed of light is c/1.4
. (It has to be a higher index of refraction than the cladding to create total internal reflection, which is the whole point of optical fibers.) Modern fibers for use over long runs do try to keep their index of refraction down as low as possible for this reason.
Use traceroute
or tracepath
to find out the network path your packets take.
add a comment |
You seem to be making two fundamental assumptions
- A domain name will always map to the same IP address.
- An IP address will always route to the same server.
Neither of these assumptions is nessacerally true. DNS servers can return different results and IP addresses can route to different servers depending on the location of the client.
As you say there is a physical limit to how low the round trip times to a distant server can be, the inescapable conclusion then if you see low ping times to what you thought was a distant server is that the server was not as distant as you thought it was.
Operators of major sites put substantial effort into optimising the locations they serve end-user traffic from, motivated by some combination of performance and cost. They may do this in-house, they may engage the services of a third party content distribution network or they may use some combination of the two strategies.
add a comment |
You can use the online tool maplatency to get a comprehensive map of ping times from your location (among other things).
Here is an example of ping times from Paris:
Interesting tool but seems not working :(
– Satish
May 20 at 20:36
add a comment |
Result from all over the world using - https://tools.keycdn.com/ping
its pure CDN base domain.
add a comment |
Your Answer
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "496"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);
StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});
function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: false,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: null,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fnetworkengineering.stackexchange.com%2fquestions%2f59138%2fnetwork-latencies-between-opposite-ends-of-the-earth%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
6 Answers
6
active
oldest
votes
6 Answers
6
active
oldest
votes
active
oldest
votes
active
oldest
votes
Google in particular uses distributed datacenters around the globe.
They announce the same IP network at various places and due to the way routing protocols work, you reach the nearest one.bbc.co.uk points to an IP address that belongs to Fastly, Inc, a content delivery network, that also has points of presence around the world, including Asia, but I don't know if they use the same technique.
(From France I have around 60ms to bbc.co.uk (151.101.192.81))
However, due to the extremely low latency you see, my bet is that you are not contacting the actual servers but this is a proxy that responds to you.
Proxy or load-balancer, which we use a lot based on which country you come from.
– user56700
May 14 at 13:06
2
It doesn't have to be a strange proxy server; it's normal to have 4-5 ms latencies to a CDN, if the ISP directly peers with that CDN and your own uplink isn't Wi-Fi or ADSL...
– grawity
May 14 at 16:31
add a comment |
Google in particular uses distributed datacenters around the globe.
They announce the same IP network at various places and due to the way routing protocols work, you reach the nearest one.bbc.co.uk points to an IP address that belongs to Fastly, Inc, a content delivery network, that also has points of presence around the world, including Asia, but I don't know if they use the same technique.
(From France I have around 60ms to bbc.co.uk (151.101.192.81))
However, due to the extremely low latency you see, my bet is that you are not contacting the actual servers but this is a proxy that responds to you.
Proxy or load-balancer, which we use a lot based on which country you come from.
– user56700
May 14 at 13:06
2
It doesn't have to be a strange proxy server; it's normal to have 4-5 ms latencies to a CDN, if the ISP directly peers with that CDN and your own uplink isn't Wi-Fi or ADSL...
– grawity
May 14 at 16:31
add a comment |
Google in particular uses distributed datacenters around the globe.
They announce the same IP network at various places and due to the way routing protocols work, you reach the nearest one.bbc.co.uk points to an IP address that belongs to Fastly, Inc, a content delivery network, that also has points of presence around the world, including Asia, but I don't know if they use the same technique.
(From France I have around 60ms to bbc.co.uk (151.101.192.81))
However, due to the extremely low latency you see, my bet is that you are not contacting the actual servers but this is a proxy that responds to you.
Google in particular uses distributed datacenters around the globe.
They announce the same IP network at various places and due to the way routing protocols work, you reach the nearest one.bbc.co.uk points to an IP address that belongs to Fastly, Inc, a content delivery network, that also has points of presence around the world, including Asia, but I don't know if they use the same technique.
(From France I have around 60ms to bbc.co.uk (151.101.192.81))
However, due to the extremely low latency you see, my bet is that you are not contacting the actual servers but this is a proxy that responds to you.
edited May 14 at 14:03
Ron Trunk
43.2k34090
43.2k34090
answered May 14 at 13:02
JFLJFL
13.1k11443
13.1k11443
Proxy or load-balancer, which we use a lot based on which country you come from.
– user56700
May 14 at 13:06
2
It doesn't have to be a strange proxy server; it's normal to have 4-5 ms latencies to a CDN, if the ISP directly peers with that CDN and your own uplink isn't Wi-Fi or ADSL...
– grawity
May 14 at 16:31
add a comment |
Proxy or load-balancer, which we use a lot based on which country you come from.
– user56700
May 14 at 13:06
2
It doesn't have to be a strange proxy server; it's normal to have 4-5 ms latencies to a CDN, if the ISP directly peers with that CDN and your own uplink isn't Wi-Fi or ADSL...
– grawity
May 14 at 16:31
Proxy or load-balancer, which we use a lot based on which country you come from.
– user56700
May 14 at 13:06
Proxy or load-balancer, which we use a lot based on which country you come from.
– user56700
May 14 at 13:06
2
2
It doesn't have to be a strange proxy server; it's normal to have 4-5 ms latencies to a CDN, if the ISP directly peers with that CDN and your own uplink isn't Wi-Fi or ADSL...
– grawity
May 14 at 16:31
It doesn't have to be a strange proxy server; it's normal to have 4-5 ms latencies to a CDN, if the ISP directly peers with that CDN and your own uplink isn't Wi-Fi or ADSL...
– grawity
May 14 at 16:31
add a comment |
A cable half-way around the globe has a minimum latency of 100 ms, 200 ms round-trip (20,000 km distance / 200,000 km/s signal speed). In reality, links aren't as the crow flies (at all) and there are additional, active components in between, adding to latency - a more realistic figure is 200 or even 300 ms one-way.
Anything with lower latency or ping-time is closer to you. As Ron's already pointed out, large service provider use globally distributed infrastructure with location-specific DNS resolution or anycasting to connect you to a nearby server.
2
As a (small sample size) test, Wellington-Madrid (very close to opposite) is about 300ms round trip pic.nperf.com/r/3190596119915611-Bc4TJSUS.png pic.nperf.com/r/3190596597699589-shrThPRn.png
– Someone Somewhere
May 15 at 8:21
add a comment |
A cable half-way around the globe has a minimum latency of 100 ms, 200 ms round-trip (20,000 km distance / 200,000 km/s signal speed). In reality, links aren't as the crow flies (at all) and there are additional, active components in between, adding to latency - a more realistic figure is 200 or even 300 ms one-way.
Anything with lower latency or ping-time is closer to you. As Ron's already pointed out, large service provider use globally distributed infrastructure with location-specific DNS resolution or anycasting to connect you to a nearby server.
2
As a (small sample size) test, Wellington-Madrid (very close to opposite) is about 300ms round trip pic.nperf.com/r/3190596119915611-Bc4TJSUS.png pic.nperf.com/r/3190596597699589-shrThPRn.png
– Someone Somewhere
May 15 at 8:21
add a comment |
A cable half-way around the globe has a minimum latency of 100 ms, 200 ms round-trip (20,000 km distance / 200,000 km/s signal speed). In reality, links aren't as the crow flies (at all) and there are additional, active components in between, adding to latency - a more realistic figure is 200 or even 300 ms one-way.
Anything with lower latency or ping-time is closer to you. As Ron's already pointed out, large service provider use globally distributed infrastructure with location-specific DNS resolution or anycasting to connect you to a nearby server.
A cable half-way around the globe has a minimum latency of 100 ms, 200 ms round-trip (20,000 km distance / 200,000 km/s signal speed). In reality, links aren't as the crow flies (at all) and there are additional, active components in between, adding to latency - a more realistic figure is 200 or even 300 ms one-way.
Anything with lower latency or ping-time is closer to you. As Ron's already pointed out, large service provider use globally distributed infrastructure with location-specific DNS resolution or anycasting to connect you to a nearby server.
answered May 14 at 17:11
Zac67Zac67
36.2k22672
36.2k22672
2
As a (small sample size) test, Wellington-Madrid (very close to opposite) is about 300ms round trip pic.nperf.com/r/3190596119915611-Bc4TJSUS.png pic.nperf.com/r/3190596597699589-shrThPRn.png
– Someone Somewhere
May 15 at 8:21
add a comment |
2
As a (small sample size) test, Wellington-Madrid (very close to opposite) is about 300ms round trip pic.nperf.com/r/3190596119915611-Bc4TJSUS.png pic.nperf.com/r/3190596597699589-shrThPRn.png
– Someone Somewhere
May 15 at 8:21
2
2
As a (small sample size) test, Wellington-Madrid (very close to opposite) is about 300ms round trip pic.nperf.com/r/3190596119915611-Bc4TJSUS.png pic.nperf.com/r/3190596597699589-shrThPRn.png
– Someone Somewhere
May 15 at 8:21
As a (small sample size) test, Wellington-Madrid (very close to opposite) is about 300ms round trip pic.nperf.com/r/3190596119915611-Bc4TJSUS.png pic.nperf.com/r/3190596597699589-shrThPRn.png
– Someone Somewhere
May 15 at 8:21
add a comment |
Major domains have servers around the world, and DNS set up so that you get an IP address that's close to you.
You can avoid this by pinging a specific host. For example, the Debian (GNU/Linux) project has a list of mirrors in countries world-wide. Most of those are specific hosts that will look up to the same IP regardless of where you are. Especially university software mirrors are almost never behind a content-delivery proxy network.
For example, mirror.csclub.uwaterloo.ca
is in Waterloo, Ontario, Canada, and does respond to ping requests. I get ~37ms ping time to it right now (early morning) from Halifax, NS, Canada.
mirror.aarnet.edu.au in Australia is probably one of the farthest away from me; Australia as a whole is network-wise far from most of the rest of the world. (although South-East Asia is closer than most to it) My ping time is ~295ms.
And yes, speed-of-light delays imposed by distance around the earth's circumference are a major part of that, as well as routing delays on hops. (And remember, it's speed of light in glass fiber not vacuum. The index of refraction of the core of an optical fiber is often something like 1.3 to 1.4, so speed of light is c/1.4
. (It has to be a higher index of refraction than the cladding to create total internal reflection, which is the whole point of optical fibers.) Modern fibers for use over long runs do try to keep their index of refraction down as low as possible for this reason.
Use traceroute
or tracepath
to find out the network path your packets take.
add a comment |
Major domains have servers around the world, and DNS set up so that you get an IP address that's close to you.
You can avoid this by pinging a specific host. For example, the Debian (GNU/Linux) project has a list of mirrors in countries world-wide. Most of those are specific hosts that will look up to the same IP regardless of where you are. Especially university software mirrors are almost never behind a content-delivery proxy network.
For example, mirror.csclub.uwaterloo.ca
is in Waterloo, Ontario, Canada, and does respond to ping requests. I get ~37ms ping time to it right now (early morning) from Halifax, NS, Canada.
mirror.aarnet.edu.au in Australia is probably one of the farthest away from me; Australia as a whole is network-wise far from most of the rest of the world. (although South-East Asia is closer than most to it) My ping time is ~295ms.
And yes, speed-of-light delays imposed by distance around the earth's circumference are a major part of that, as well as routing delays on hops. (And remember, it's speed of light in glass fiber not vacuum. The index of refraction of the core of an optical fiber is often something like 1.3 to 1.4, so speed of light is c/1.4
. (It has to be a higher index of refraction than the cladding to create total internal reflection, which is the whole point of optical fibers.) Modern fibers for use over long runs do try to keep their index of refraction down as low as possible for this reason.
Use traceroute
or tracepath
to find out the network path your packets take.
add a comment |
Major domains have servers around the world, and DNS set up so that you get an IP address that's close to you.
You can avoid this by pinging a specific host. For example, the Debian (GNU/Linux) project has a list of mirrors in countries world-wide. Most of those are specific hosts that will look up to the same IP regardless of where you are. Especially university software mirrors are almost never behind a content-delivery proxy network.
For example, mirror.csclub.uwaterloo.ca
is in Waterloo, Ontario, Canada, and does respond to ping requests. I get ~37ms ping time to it right now (early morning) from Halifax, NS, Canada.
mirror.aarnet.edu.au in Australia is probably one of the farthest away from me; Australia as a whole is network-wise far from most of the rest of the world. (although South-East Asia is closer than most to it) My ping time is ~295ms.
And yes, speed-of-light delays imposed by distance around the earth's circumference are a major part of that, as well as routing delays on hops. (And remember, it's speed of light in glass fiber not vacuum. The index of refraction of the core of an optical fiber is often something like 1.3 to 1.4, so speed of light is c/1.4
. (It has to be a higher index of refraction than the cladding to create total internal reflection, which is the whole point of optical fibers.) Modern fibers for use over long runs do try to keep their index of refraction down as low as possible for this reason.
Use traceroute
or tracepath
to find out the network path your packets take.
Major domains have servers around the world, and DNS set up so that you get an IP address that's close to you.
You can avoid this by pinging a specific host. For example, the Debian (GNU/Linux) project has a list of mirrors in countries world-wide. Most of those are specific hosts that will look up to the same IP regardless of where you are. Especially university software mirrors are almost never behind a content-delivery proxy network.
For example, mirror.csclub.uwaterloo.ca
is in Waterloo, Ontario, Canada, and does respond to ping requests. I get ~37ms ping time to it right now (early morning) from Halifax, NS, Canada.
mirror.aarnet.edu.au in Australia is probably one of the farthest away from me; Australia as a whole is network-wise far from most of the rest of the world. (although South-East Asia is closer than most to it) My ping time is ~295ms.
And yes, speed-of-light delays imposed by distance around the earth's circumference are a major part of that, as well as routing delays on hops. (And remember, it's speed of light in glass fiber not vacuum. The index of refraction of the core of an optical fiber is often something like 1.3 to 1.4, so speed of light is c/1.4
. (It has to be a higher index of refraction than the cladding to create total internal reflection, which is the whole point of optical fibers.) Modern fibers for use over long runs do try to keep their index of refraction down as low as possible for this reason.
Use traceroute
or tracepath
to find out the network path your packets take.
answered May 15 at 10:08
Peter CordesPeter Cordes
1213
1213
add a comment |
add a comment |
You seem to be making two fundamental assumptions
- A domain name will always map to the same IP address.
- An IP address will always route to the same server.
Neither of these assumptions is nessacerally true. DNS servers can return different results and IP addresses can route to different servers depending on the location of the client.
As you say there is a physical limit to how low the round trip times to a distant server can be, the inescapable conclusion then if you see low ping times to what you thought was a distant server is that the server was not as distant as you thought it was.
Operators of major sites put substantial effort into optimising the locations they serve end-user traffic from, motivated by some combination of performance and cost. They may do this in-house, they may engage the services of a third party content distribution network or they may use some combination of the two strategies.
add a comment |
You seem to be making two fundamental assumptions
- A domain name will always map to the same IP address.
- An IP address will always route to the same server.
Neither of these assumptions is nessacerally true. DNS servers can return different results and IP addresses can route to different servers depending on the location of the client.
As you say there is a physical limit to how low the round trip times to a distant server can be, the inescapable conclusion then if you see low ping times to what you thought was a distant server is that the server was not as distant as you thought it was.
Operators of major sites put substantial effort into optimising the locations they serve end-user traffic from, motivated by some combination of performance and cost. They may do this in-house, they may engage the services of a third party content distribution network or they may use some combination of the two strategies.
add a comment |
You seem to be making two fundamental assumptions
- A domain name will always map to the same IP address.
- An IP address will always route to the same server.
Neither of these assumptions is nessacerally true. DNS servers can return different results and IP addresses can route to different servers depending on the location of the client.
As you say there is a physical limit to how low the round trip times to a distant server can be, the inescapable conclusion then if you see low ping times to what you thought was a distant server is that the server was not as distant as you thought it was.
Operators of major sites put substantial effort into optimising the locations they serve end-user traffic from, motivated by some combination of performance and cost. They may do this in-house, they may engage the services of a third party content distribution network or they may use some combination of the two strategies.
You seem to be making two fundamental assumptions
- A domain name will always map to the same IP address.
- An IP address will always route to the same server.
Neither of these assumptions is nessacerally true. DNS servers can return different results and IP addresses can route to different servers depending on the location of the client.
As you say there is a physical limit to how low the round trip times to a distant server can be, the inescapable conclusion then if you see low ping times to what you thought was a distant server is that the server was not as distant as you thought it was.
Operators of major sites put substantial effort into optimising the locations they serve end-user traffic from, motivated by some combination of performance and cost. They may do this in-house, they may engage the services of a third party content distribution network or they may use some combination of the two strategies.
edited May 15 at 16:38
answered May 15 at 16:32
Peter GreenPeter Green
8,21221230
8,21221230
add a comment |
add a comment |
You can use the online tool maplatency to get a comprehensive map of ping times from your location (among other things).
Here is an example of ping times from Paris:
Interesting tool but seems not working :(
– Satish
May 20 at 20:36
add a comment |
You can use the online tool maplatency to get a comprehensive map of ping times from your location (among other things).
Here is an example of ping times from Paris:
Interesting tool but seems not working :(
– Satish
May 20 at 20:36
add a comment |
You can use the online tool maplatency to get a comprehensive map of ping times from your location (among other things).
Here is an example of ping times from Paris:
You can use the online tool maplatency to get a comprehensive map of ping times from your location (among other things).
Here is an example of ping times from Paris:
answered May 16 at 8:56
GohuGohu
141114
141114
Interesting tool but seems not working :(
– Satish
May 20 at 20:36
add a comment |
Interesting tool but seems not working :(
– Satish
May 20 at 20:36
Interesting tool but seems not working :(
– Satish
May 20 at 20:36
Interesting tool but seems not working :(
– Satish
May 20 at 20:36
add a comment |
Result from all over the world using - https://tools.keycdn.com/ping
its pure CDN base domain.
add a comment |
Result from all over the world using - https://tools.keycdn.com/ping
its pure CDN base domain.
add a comment |
Result from all over the world using - https://tools.keycdn.com/ping
its pure CDN base domain.
Result from all over the world using - https://tools.keycdn.com/ping
its pure CDN base domain.
answered May 20 at 20:41
SatishSatish
2,02012765
2,02012765
add a comment |
add a comment |
Thanks for contributing an answer to Network Engineering Stack Exchange!
- Please be sure to answer the question. Provide details and share your research!
But avoid …
- Asking for help, clarification, or responding to other answers.
- Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fnetworkengineering.stackexchange.com%2fquestions%2f59138%2fnetwork-latencies-between-opposite-ends-of-the-earth%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
Required, but never shown
I've run tests that don't involve CDNs and I'd estimate you're more likely to have latencies of 250 - 500ms between you and servers actually in Europe. Perhaps less, but you're right to think that global distances have effects on minimum latencies that no network topology can reduce.
– Todd Wilcox
May 14 at 16:15
Why don't you provide traceroute full output?
– Satish
May 20 at 20:31