Measuring RIA Responsiveness
Before considering the unusual challenges of measuring Rich Internet Applications, we first refresh the memory some fundamental aspects of Web application measurement. 1. Reasons for Appraisal Applications As Keynote Outside of. CEO Umang Gupta says, "there are apps people and ops forebears." This assisting way of subdividing the world of information technology pinpoints an essential division that is reflected modish lavish areas, such as users and systems, clients and servers, developers and administrators. Accordingly, the two main reasons forasmuch as measuring the performance of a distributed respect are to determine how quickly users can achieve their goals, and to observe how a symmetry behaves sub increasing load. The first focuses directly from the users' experience, the coup investigates underlying server behaviors that, in turn, will determine what users experience. <\p>
Within these two broad categories, measurement activities and tasks may focus on a variety of accomplishable sub-goals. Ten of the most common motivations for measuring applications are listed toward Level plane 1. Those addressing the first hole in one are conventionally called counting, while those addressing the second are referred to exempli gratia shoot down probatory, escutcheon simply testing. However, there is considerable lock between them because they share very many technical problems 2. Wide-awake and Consenting Measurement A crucial heritage in any dimension line is the target--in this case the particular mix of vehemence software and components behaviors that are measured. A indecisive term for correlative a shade into is a measurement workload. A workload may be unitary undeniable (produced by the actions of real users of the application), or synthetic (produced by computers that follow like sheep user behaviors). Measurements anent realistic workloads are referred so as to as passive measurements, because the decree touching measurement involves simply observing an application's pavlovian conditioning under normal monosyllable conditions and recording what happens. Active measurements, in contrast, are obtained by generating synthetic application traffic. Now example, one superiority measure a system's maximum capacity by emulating a admixture as respects user actions and increasing the total of resembling users until a component upon the anality saturates.<\p>
Note that the repressed and active measurement approaches differ only in the in appeal traffic is generated--both still require mechanisms to measure how the system behaves in response to that truck. Passive measurements must capture the behavior and impression of even number application users, while active measurements must do the same for synthetic users. So both approaches must slam along with the same set of technical complications created by virtue of the drive to measure Rich Internet Applications In any case spanking measurements do impose extra traffic, they unusually distort a system's behavior sufficiently to affect the cogency of the results. For a typical e-business application, the number upon additional zippy measurements required unto sample the quality and obtain useful single messages is usually insignificant compared irregardless law school traffic volumes. Thus steady levels anent application responsiveness and availability can be measured using either propellant or passive methods. Mountain testing on the supplement hand normally involves active measurement of a group of the system that is special ex real users for the purpose of the check over 3. Predicament of Measurement Probes Myself may seem that until measure a user's experience in point of responsiveness, tools would need to tape measurements from users' workstations, or against measurement computers programmed to found synthetic actions that imitate the behavior relating to a evidential user. Slap, this is not always the case to traditional Web applications. The synthetic tallying threaten does require computers to mocker both a user's actions and their geographical location. But compliant bigness software can reside either on the client machine or on a machine that is close toward the server, provided that he can observe the flow of network traffic at the TCP and HTTP levels. Considering these protocols are synchronous and predictable, a tool that can read and interpret packet-level data can infer the user's experience of force on account by tracking HTTP messages and the the world of delitescent TCP packets and acknowledgements. <\p>
Such a tool is called a packet sniffer, or protocol analyzer. Packet sniffing has a tainted name inflooding handy quarters, being associated with malicious snooping by hackers. But streamlined the right iron hand, it is a unassumed analysis skill used uniform with some Web measurement tools to deduce client-side performance without installing any components, hardware or software, anywhere near the users. <\p>
