End-user experience traffic data capture and segmentation

You might deploy various methods to capture traffic so you can monitor the end-user experience with a web application:

Network-based packet capture, which monitors browser-based events as they traverse the network in real time
Synthetic transaction generation to predict the end-user experience with a web application
Client-side agents for cloud-based services
Client-side endpoint instrumentation for Rich Internet Applications (RIAs) or Content Distributed Networks (Akamai, for example)

Traffic capture methods

The following topics summarize the methods you use to capture traffic.

Network-based traffic capture with a Collector and Analyzer

A BMC Real End User Experience Monitoring system consists of at least one Real User Collector device supplying data to at least one Real User Analyzer component. Alternatively, you can deploy a hardware or software version of the Real User Monitor, which provides a combined Real User Analyzer and Real User Collector.

Real User Collector components capture traffic passing between your web applications and end users via a network tap or mirror port on a switch or load balancer. Traffic inclusion and exclusion policies allow you to control what application data is captured. Using uploaded SSL keys, the Real User Collector device can decipher encrypted traffic. It can also obfuscate or delete private data. You can monitor the flow of traffic into the device via status information on its Home page and the traffic capture statistics page.
Real User Analyzer components continuously retrieve data from one or more Real User Collector devices. You can control what data the Real User Analyzer device consumes from a given Real User Collector device by setting filters for each Collector feed. You can prioritize feeds so that the most important data is available to dashlets, reports, and other monitoring features of BMC Real End User Experience Monitoring.

For deployment options, see BMC-Real-End-User-Experience-Monitoring-deployment-use-cases. For details about traffic capture technologies, see Traffic-capture-and-tapping-points. For more information about filters, see Establishing-traffic-data-feeds-between-an-Analyzer-and-a-Collector.

Traffic capture in a cloud environment

The Real User Cloud Probe is a software-based agent that you can install directly onto a Windows or Linux OS to provide remote real end-user monitoring capabilities in environments where network taps or spanning ports are not accessible. The Cloud Probe collects performance data directly from the network interface of its host and forwards it to a standard Real User Collector component.

With this capability, BMC Real End User Experience Monitoring can monitor applications that are deployed anywhere, whether in a public cloud environment, such as Amazon EC2, or in a private data center. The product can also support monitoring for applications with hybrid deployment models, leveraging a combination of private data-center infrastructure and cloud-based infrastructure.

The Cloud Probe service can capture only HTTP or HTTPS traffic destined for the system on which it is hosted; it cannot capture traffic going out of the system.

For deployment options, see Cloud-Probe-deployment-use-cases.

Traffic capture in a Content Delivery Network

BMC Real End User Experience Monitoring enables you to set up monitoring for web applications with a Content Delivery Network (CDN) service. A CDN is a large, distributed system of servers deployed in multiple data centers on the internet or on private networks. The browser retrieves content for the web application from a variety of servers. Because cached content does not always come from an origin server (where the web application is stored), deriving performance statistics requires JavaScript code to be embedded in your web application pages. You deploy a page-render beacon (or a web beacon) to calculate the page-render time metric for content served by a CDN.

For more information, see How-performance-metrics-are-calculated-in-a-Content-Delivery-Network.

Traffic capture for a Rich Internet Application

The RIA Visibility toolkit enables you to measure the performance of applications that manage user interaction, display content without communicating with the server (for example, using Flash and Flex technology), and use the web primarily for data transfer. Such an application is sometimes known as a rich internet application (RIA).

For more information, see Measuring Performance for Rich Internet Applications.

Synthetic traffic generation

BMC Synthetic End User Experience Monitoring uses prerecorded .ltz scripts to simulate end-user transactions. The product determines the health of business applications and measures their response times by running recorded transactions at key locations throughout the enterprise or on the internet. BMC provides some basic scripts, and you can create realistic and customizable scripts through an external scripting tool. The scripts require a compatible run-time environment.

The TEA Agent sends data over HTTPS to a Real User Collector that supplies data to a Real User Analyzer. You establish the component association, including selecting which Real User Collector to use, through the TEA Agent policy. For synthetic transactions, there is no need to to use the capturing interface on the Real User Collector.

For more information, see BMC-Synthetic-End-User-Experience-Monitoring-overview.

Traffic segmentation with Watchpoints

Traffic capture can result in an overwhelming amount of information. To make it easier to monitor only the parts of your web traffic that interest you, you can define precise segments of web traffic to monitor, known as Watchpoints. Following are some examples of traffic segments for which you can define Watchpoints:

Traffic to a particular web application
Traffic from a particular group of end users
Traffic from a particular geographic region
Traffic involving a particular part of your infrastructure
Traffic from a particular client platform

For each Watchpoint, the system aggregates traffic volume, availability, and performance metrics in 5-minute intervals.