The Real User Analyzer component is sized according to the following table:
Real User Analyzer sizes
|Expected volume of traffic|
|Demo mode||2 vCPU||4 GB||20 GB||NA for Demo mode|
|900 to 2,000 hits per second|
|1,500 to 3,500 hits per second|
|2,500 to 7,500 hits per second|
Real User Collectors feed their data to Real User Analyzer instances, and you must size these instances to ensure that they can handle the generated load. As more Collector instances are added to the system, you must re-evaluate Analyzer capacity to ensure that the increased load is not a problem.
Sizing for traffic rate
The size that you choose to deploy depends on the volume of traffic:
- Low — 900 to 2,000 hits per second
- Medium — 1,500 to 3,500 hits per second
- High — 2,500 to 7,500 hits per second
To understand the ability of an Analyzer instance to handle the traffic load, monitor the Home page of the Collector instances that are feeding it and check the sampling rate, as shown in the following example page.
Real User Collector Home page
The sampling rate represents the number of hits per second that are being discarded because the target Analyzer cannot accommodate the load, causing the Analyzer to take more time than is acceptable to pull its data from the Collector.
Sizing for concurrent pages and sessions
The size that you choose to deploy might depend on the number of concurrent sessions or pages that an Analyzer instance must process:
- Low — 50,000 pages or 150,000 sessions
- Medium — 70,000 pages or 210,000 sessions
- High — 90,000 pages or 275,000 sessions
Increased traffic volume usually correlates to more users requesting more web pages. However, a given site can have a high traffic rate and a low number of users (or vice versa). So, independent of the traffic volume, you might have to adjust the resource allocation of an Analyzer instance to accommodate page and session-tracking needs.
To "rebuild" pages and sessions, an Analyzer instance uses memory caches, the size of which determines how many simultaneous page requests and simultaneously active users that the system can track.