Software performance

There are different ways to track and analyze statistics.

Some of them require less server resources, but will also severely limit how you can work with the statistics (only a small number of pre-defined reports is available, the data is not available in real time). More flexible approaches (real-time data, ability to create your own reports based on the gathered data) usually need more server resources too. Generally, the more flexible the statistics software is — the more server resources it will use to track and analyze the data.

Initially, when we've only started the work on Stuffed Tracker, we've decided that the product will provide as much data as possible in real-time and that it will have very flexible tools for statistics analysis. Inevitably, this also increased the amount of server resources that Stuffed Tracker uses to perform its tasks.

Usually, Stuffed Tracker works on one server with only one database, and quite often it shares the server resources with lots of other software which runs on the same server. In such scenario, the performance of the tracker might suffer, especially if it is used to process a lot of traffic (sites with a lot of visitors, high traffic ad campaigns).

We've spent more then a year to develop such tools as Caching, Archiving, Statistics removal and Delayed tracking — all of them help to dramatically increase the tracker's performance when working with large sets of statistical data. In the future we also plan to add load balancing functionality to make it possible to distribute the load between multiple servers and databases.

We highly recommend using the trial version of the product in order to assess the tracker's performance on your server with your traffic volume.