For modern big and difficult software systems, a lot of tasks are performed to implement multiple purposes. Various testing methods can be used and integrated, and a lot of resources are involved as well.
System testing performs software testing of the general system operations as a whole, commonly from a client's point of view.
The main relation is how the software system operates as a whole under the operational environment of actual clients. Consequently, in system testing the whole system is treated as a black-box, where external functions are tested.
The client's point of view adopted for this testing stage. That is why usage-based statistical testing techniques are usually used.
Because of this relation and the point of view, the items to be tested stay at really high levels of abstraction. For instance, high-level functions or elements may be tested, commonly those straightly visible to the clients, but not execution details or those components far removed from clients.
The most wide-spread methods comprise:
1. Finite-State Machines
Finite-state machines for the system functions may be created for more systematic testing than only using the checklists. Every state here represents a main function anticipated by and visible to target clients. Such testing models and methods are black-box ones, because they are based on external functions instead of executions. Different coverage criteria may be used as stopping criteria.
2. High-Level Functional Checklists
They are usually used to assure that all the main functions anticipated by the clients are present and satisfy all the clients' needs. We can try to reach full coverage of all main functions, consequently resulting in coverage-based testing. We can reach reliability purposes by focusing functions weighty to and often used by clients, resulting in usage-based statistical testing.
For embedded software systems or for miscellaneous systems with significant software elements, such as software controlled medical equipment and modern telecommunication networks, the term "system" commonly means the entire system with the software part as an element or a sub-system.
In such type of systems, system testing takes on supplementary implication as well, which we submit to as super-system testing.
Integration testing and system testing can be applied to such super-system testing, to examine for interface, interplay and interoperability troubles among various sub-systems, and also examine the general super-system operations.
If we are talking about software testing company http://qatestlab.com/company/about-us/ and its software testers http://qatestlab.com/company/our-team/, knowledge about general product functions, application domain and market segment, and client anticipations and their usage of the system is weightier than product execution points.
Author Resource:-
QATestLab is independent,offshore software testing company located in Kiev, Ukraine. QATestLab performs testing of products on each stage of software development cycle.
![[Valid RSS feed]](http://articlepile.com/images/icon_Rss.png "XML Feed For RSS"){kind=icon}
Category Rss Feed
Print This Article
Add To Favorites