Recovery Model for Survivable System through Resource Reconfiguration

A survivable system is able to fulfil its mission in a timely manner, in the presence of
attacks, failures, or accidents. It has been realized that it is not always possible to anticipate
every type of attack or failure or accident in a system, and to predict and protect against those
threats. Consequently, recovering back from any damage caused by threats becomes an
important attention to be taken into account. This research proposed another recovery model
to enhance system survivability. The model focuses on how to preserve the system and
resume its critical service while incident occurs by reconfiguring the damaged critical service
resources based on available resources without affecting the stability and functioning of the
system. There are three critical requisite conditions in this recovery model: the number of
pre-empted non-critical service resources, the response time of resource allocation, and the
cost of reconfiguration, which are used in some scenarios to find and re-allocate the available
resource for the reconfiguration. A brief specifications using Z language are also explored as
a preliminary proof before the implementation .. To validate the viability of the approach,
two instance cases studies of real-time system, delivery units of post office and computer
system of a company, are provided in ensuring the durative running of critical service. The
adoption of fault-tolerance and survivability using redundancy re-allocation in this recovery
model is discussed from a new perspective. Compared to the closest work done by other
researchers, it is shown that the model can solve not only single fault and can reconfigure the
damage resource with minimum disruption to other services.