Conception and implementation of a modular simulator for the adaptive-dynamic replication framework adGSV
Conception and implementation of a modular simulator for the adaptive-dynamic replication framework adGSV
Master's thesis or Diplom thesis
| Contact person: | Prof. Dr.-Ing. Oliver Theel | |
| In progress since 22 April 2008 by Thorsten Simon | ||
Background
Data replication is used to minimise the possibility of access failures to urgently required critical data (high access availability), but also to reduce access times to this data (load balancing).
A prototype for consistent data replication was created in the department. This prototype includes a component that allows any so-called Coterie-based replication procedure to be specified. The specification of a desired special replication strategy is carried out by manually specifying an acyclic graph (in text form) or automatically by a generator. The prototype is able to process both static and dynamic replication strategies. Dynamic replication strategies are generally characterised by higher access costs on the one hand and higher access availability of the replicated data on the other.
At design time, the replication strategy to be used initially is determined by the administrator. In principle, the replication strategy used can be adapted to changing requirements at runtime, e.g. by adding and removing computers participating in the replication process or changing the replication strategy.
Discrete, event-driven simulation is used to determine the properties of a selected strategy, such as its access costs and access availability. Comparative measurements of several strategies are used to determine the best strategy for a given deployment scenario.
Job description
The task to be solved as part of the thesis is the design, implementation and integration of a modular simulator into the existing prototype. The simulator must be able to control the replication framework accordingly, execute the simulation and then analyse the collected data. By modularising the individual components of the simulator, the aim is to create the flexibility to adapt the simulator to different scenarios simply by reconfiguring it. Furthermore, it should be possible to define the simulation process or the sequence of actions of a simulation in fine granularity. The existing simulator can be used as a basis.
Finally, a specific deployment scenario is to be evaluated using the simulator in order to then design a new and specifically better strategy for this scenario and then also evaluate it.