Design of Scenario Based Application optimized Data Replication Strategies through Genetic Programming


Data replication plays a very important role in distributed systems because a single replica is prone to failure, which is devastating for the availability of the access operations. High availability and low cost for the access operations as well as maintaining data consistency are major challenges for reliable services. Since failures are often inevitable in a distributed paradigm thereby greatly affecting the availability of services. Replication masks failures in order to achieve a fault-tolerant distributed environment. Data replication is, hence, an appropriate means to provide highly available data access operations at relatively low operation costs. Although there are several contemporary data replication strategies being used, the question still stands which strategy is the best for a given scenario or application class assuming a certain workload, its distribution across a network, availability of the individual replicas, and cost of the access operations. In this regard, the research to be presented focuses on analysis, simulation, and machine learning approaches,
particularly genetic programming, to automatically identify and design such replication strategies that are optimized for a given application-scenario, based on predefined constraints and properties exploiting a so-called voting structure of directed acyclic graphs. Genetic programming explores new unknown replication strategies. It evolves the population of replication strategies (representing each a computer program) gradually but consistently to eventually meet the desired criteria. Furthermore, the research introduces strong multi-crossover and multi-mutation operators which strengthen our machine learning framework, at the same time guaranteeing consistency of the solutions, to generate innovative hybrid replication strategies.
Betreuer: Prof. Dr. Oliver Theel


20. Januar 2020, 16:15


Syed Mohtashim Abbas Bokhari Universität Oldenburg


OFFIS, Escherweg 2, Raum F 02

Webmas1pdstleer (marg+aco.grawygunder+gzk@u2l3ol.dspes8gi) (Stand: 07.11.2019)