Naturally, there exist many different implementations for every layer
of the system to deal with the HotSwap-originated problems.
Considering the wide variety of possible implementations, it would
be very useful if one could forecast certain quantitative key values to
evaluate the benefits and drawbacks of a particular solution. Interesting
properties are the system´s performance on the one hand and the system´s
availability on the other hand. Furthermore the system´s development-,
production-, and maintenance-costs have to be taken into consideration
when choosing an implementation.
Because it is of course not possible to test a system in its
design phase, answers must be obtained from building models of the particular
implementations. These models can be evaluated with formal methods or simulated.
Because of the complexity of the inspected systems, evaluating these models
is only possible using convenient computer-aided tools. We currently investigate
which formal methods are applicable for the evaluation of HotSwap-based
systems and which tools may be used to evaluate/simulate these models. |
