Test Server Methodology

On-net and Off-net Test Servers

Test servers may be deployed inside a customer’s own private network (“on-net” test servers). This allows them to segment end-to-end network performance and determine the performance of their own network versus third-party networks. For example, an ISP (internet service provider) can see what impact third party networks have on their end-users’ network layer tests by placing test servers within their own network and at major national and international peering locations.

It is also important that measurements collected by the test architecture support the comparison of “off-net” network performance, in an unbiased manner. Off-net test servers are located outside of the bounds of any one organization’s private network, and seek to provide a neutral and consistent measurement target. To this end, we maintain a fleet of off-net public test servers deployed at major internet exchanges and peering locations globally. Test servers are carefully sited to ensure optimal connectivity on a market-by-market basis. Measurements made against off-net servers represent "real world" end-user experiences, and off-net servers can also help bring to light peering problems or routing issues.

Regulator reports use data collected from measurements against off-net servers, and customers who use off-net servers can ensure that they are getting a realistic overview of their customers' experience. Testing to multiple servers in a variety of locations helps build a picture of network health beyond customers’ immediate borders, allowing them to view local, national, and international performance to better understand the service they are providing to their users.

Test Server Selection

Our Device Agents (routers and mobile apps) select the server based on one of the following criteria:

• Nearest off-net test server – by running round-trip latency checks to all nearby test servers before measurement begins.

• Sets of on-net servers – can also be setup to behave in the same way, selecting the nearest on-net test server using a ping test.

  • Note: the term “nearest” refers to the test server nearest to the agent from the point of view of network delay, which may not necessarily always be the one nearest geographically.

• "Random" off-net server – a set of servers configured for "random" selection, meaning the tests will pick a different appropriate server from the list every time they run. This is useful when all the servers in the list are in the same location with the same expected latency and performance, allowing us to automatically balance the testing load across all the servers in the list.

Alternatively, it is possible to override test server selection based on latency and implement a static configuration so that the agent will only test against a test server chosen by the administrator. This is so that the administrator can choose to test to any test server that is of interest to the specific project and to maintain configuration consistency.

Data Storage

For privacy and security purposes, no measurement data is stored on test servers. The test servers provide a “dumb” endpoint for the Device Agents to test against. All measurement performance results are recorded by the Device Agents, not the test servers, which are then transmitted from the agent to data collection servers managed by us.