Many people contact us in regards to latency (high pings) issues that they are experiencing within the BigPond network. Please be aware that there are many factors (apart from routing) that contribute to latency.
Some common factors include:
The physical distance between hops on the network (a latency result of 170-200ms would be expected when connecting to a server in the US)
Using a wireless connection to your router.
The configuration of any firewalls or antivirus programs.
Your internet connection speed (or the utilisation of the bandwidth), either on the your individual service, or in the local exchange (congestion)
Retransmissions and/or errors within the network, as each hop can increase the latency slightly during conversion and routing processes.
TCP stack / network protocol engineering
What we need to remember is that regardless of how advanced the technology is, networking is limited to the speed of electricity and light.
Any tests conducted from a ping or traceroute may be considered low priority traffic, and that missed hops or intermittent drops may not actually be indicative of a fault condition. Network devices may be set up with queuing, policing, blocking etc. This may be to protect the network from reconnaissance, denial of service or to give priority to time sensitive traffic, VOIP for example.
If you believe you are experiencing latency you can try the following basic troubleshooting:
Connect your device directly to your modem via an Ethernet cable. A direct Ethernet connection gives you a better flow of data.
Make sure no one else is using your internet connection. Kick other users off, check to see what other devices are connected and turn them off or disconnect their connection.
Check what other applications are running in the background using your bandwidth, and turn them off. Examples of these applications include Peer-2-peer/torrenting programs downloading and uploading music and video, cloud services syncing data to the cloud and Malware or virus’ and in some cases, Internet Browser Add-ons or extensions.
Other applications which may be limiting your connection are firewalls and antivirus programs. Many firewalls substantially slow down internet connection speed. Disabling firewalls that directly monitor web traffic will lower your ping. (Please be aware that doing this will open your device up to risks of contracting a virus)
Choose a server which is the geographically closest to you. The following map shows the international links Telstra Global maps, try for the shortest.
Reset your modem to factory default settings.
You could also try changing your DNS settings, try the Google public DNS server (126.96.36.199, 188.8.131.52) and see if there is any change.
Test on a different device to see if it is a device issue or a network issue.
As pings and traceroutes may be treated as low priority traffic, it is important to keep this in mind when using test tools
The internet is a network of interconnecting servers, each one is owned by various entities and providers. When reporting issues with routing, your service provider may not be able to escalate the issue directly to the owner of the affected server.
This is reflected in a similar way to reporting a broadband fault in Australia, the user needs to contact their ISP to report the fault, who then contacts the owner of the physical infrastructure to investigate and resolve. The same happens on a larger scale with routing issues.
Due to the number of organisations that the issue may need to be escalated through, it can take some time for the communication to reach the provider needing to complete the work, and may be resolved faster if reported by the provider closest to the problem server.
In the below traceroute example, a problem with routing in the AT&T network would be escalated from Telstra BigPond to Telstra Global, who will escalate to Savvis, who escalates it to AT&T. If the problem was within Savvis, then the quickest path to report this is via AT&T. You can find out more about peering here.