Latency refers to the delay or time it takes for data to travel from one point in a network to another. For example, the time difference between when a device sends a request for data and when it receives the requested data. A slow platform can be caused by high latency, which is sometimes referred to as 'lagging', e.g. my laptop seems to be lagging today.
Where can high latency occur?
Typically there are three places high latency can occur:
Upstream: This refers to the flow of data being pushed upstream, e.g. a platform user’s computer is requesting The Happiness Index sends the data from databases to their computer via the platform.
Application: This means the primary source, e.g. The Happiness Index platform.
Downstream: This happens when The Happiness Index platform is requesting data from their databases. For example, when a platform user is analysing results the platform will request data from the databases and the data will be packaged up and sent to the platform.
An example of how data flows downstream and upstream below:
Causes of high latency - Upstream
Network: If there is a high volume of network traffic, e.g. lots of employees at the same company as the platform user are using the same network or there is a high number of devices linked to the network and the network has become congested.
Computer load: A platform user's computer may have a lot of background processes running, e.g. multiple resource heavy tabs open in their browser or additional applications running on their computer which may cause delay.
Distance: If a business has employees who work remotely from home or other locations, accessing data or applications over the internet can cause high latency due to the distance between the employee and the business's network. For example, an employee accessing the platform via a VPN they use to connect to their company’s network whilst working from home could create higher latency.
Internet: The level of bandwidth available to a platform user could create high latency if the available internet speed is low.
Causes of high latency - Application
Inefficient code: The code which runs the application could be inefficient, for example, longer code means requests take longer. By optimising code to be efficient latency becomes less of an issue.
Bugs: An application which has a bug, e.g. something is stopping the application from working in the way it was designed to do, can cause latency. For example, a progress wheel keeps spinning because the application can’t load the page due to an error.
Causes of high latency - Downstream
Autoscaling: This is a cloud computing feature that scales cloud services such as server capacities up or down automatically based on situations such as traffic levels. Autoscaling may not happen quickly enough as demand peaks, or autoscaling may happen too quickly which can lead to temporary overloading. This is the most common downstream issue to cause high latency.
Third party deployments: A platform user may experience high latency due to a third party downstream provider deploying an update or new features at the time the platform user is using the platform.
How to diagnose an issue
The first thing to check is whether the issue occurs upstream. To do this there are a number of questions to ask:
- Are you experiencing internet connectivity issues?
- Are you experiencing slow internet speeds?
- Are you experiencing network issues?
If the answers to any of the questions above is yes then upstream is the most likely cause of the high latency and we recommend to look into this with your organisation's IT support.
If the answers are no, contact THI for our tech community to investigate. Please include as much information as possible when contacting e.g. the type of browser used, the site and page accessed, the error encountered, and attach any videos or screenshots you take.