Im Rahmen einer Zusammenarbeit mit Hitachi ist die folgende Masterarbeit ausgeschrieben. Bitte kontaktieren Sie mich bei Interesse unter klaus-tycho.foerster@tu-dortmund.de
Container Orchestration with Stateful Failover Support
With more than 35'000 employees, Hitachi Energy is the largest global supplier of products, systems and services for building and operating electric power infrastructure. At Hitachi Energy Research, the globally distributed research organization of Hitachi Energy, we apply interdisciplinary knowledge to invent and design the power system of the future. In close collaboration with our business units, academia, and customers, we aim to address societal challenges such as rising global energy demand, growing integration of renewable energy sources and climate change. We deliver innovations that ensure that electrical power is supplied reliably, efficiently, and effectively.
Are you looking for an M.Sc. thesis in software systems? At Hitachi Energy Research Switzerland, we have a thesis proposal on supporting stateful failover in container orchestrators such as Kubernetes.
More specifically, the default Kubernetes redundancy concept involves multiple active replicas, which share the load among each other, and are independent. As a result, when a replica fails or the node on which it is running fails, other replicas can ensure continued service availability. This mechanism is well-suited for stateless services. However, this is not the case for stateful services having an internal state that needs to be preserved. In this thesis, you will explore and evaluate the use of tailored Kubernetes operators or alternative concepts to implement an active-standby failover mechanism with the support of a checkpoint and restore tool (e.g., CRIU). The target use case for this work is a critical infrastructure scenario, in particular power grids.
Apply now for this opportunity and become a member of our team of highly skilled researchers and engineers. The expected duration for this thesis is 6 months.
Your responsibilities
· Acquire knowledge on the theory (failover, redundancy, fault tolerance) and related technology (Kubernetes, CRIU) and summarize findings.
· Design, implement and evaluate the performance of the proposed Kubernetes operator
· Report on your work, including presentation and interpretation of results
· Contribute to a scientific publication
Your background
· Currently pursuing an M.Sc. degree in Computer Science or adjacent field
· Comfortable with Linux. Experience in Linux kernel development is a plus.
· Experience with the Go programming language or willingness to learn it.
· Experience with Docker, Kubernetes.
· Fluent in spoken and written English.
· Self-driven and curious, with good organizational skills.
Mehr Details gibt es auch hier.