In the age of IoT, big data, blockchain and machine learning, we handle a massive number of devices generating huge amount of data which are stored and analyzed by advanced aglorithms in a large scale of edge and cloud computing infrastructures. What would be key foundational concepts of services computing and cloud computing that enable seamless integration and execution of such big data and complex algorithms?
The goal of this lecture is to discuss and study advanced theoretical foundations, system designs, algorithms and system analysis of recent developments regarding Service Computing and Cloud Computing in large-scale distributed systems that enable contemporary data science, IoT analytics and deep learning. We do not learn how to program IoT Cloud Systems (as shown in Advanced Services Engineering), Web services (as shown in Advanced Internet Computing) or particular distributed systems technologies (as in Distributed Systems Technologies). But we will focus on on advanced algorithms, system models, and performance monitoring and analysis techniques for state-of-the-art cloud services, software service systems, IoT, and edge computing that are essential and foundational for building today and tomorrow's reliable and high-quality data science and deep learning applications. This course aims at covering deep knowledge about services and cloud computing that are missing in other courses.
Students will learn and research state-of-the-art algorithms and techniques through literature study, individual and team development project with real-world case studies and questions.
This course presents and studies hot topics in advanced algorithms, system models, and performance monitoring and analysis techniques for state-of-the-art cloud services, software service systems, IoT, and edge computing. From the system perspective, we will focus on
- Complex and hybrid cloud computing systems, such as IoT cloud systems and federated clouds as continuum computing environments where complex applications will be developed and executed.
- Complex and big data analytics for emerging domains
Several enabling techniques are also discussed for the above-mentioned systems to support user's requirements, such as
- Advanced algorithms for cloud computing, IoT and edge computing: high availability, data sharding, geographical multi-cloud load balancing, automatic discovery and formation of container clusters
- Orchestration and elasticity control algorithms: IoT/cloud/edge coordination and control algorithms, big data pipeline orchestration
- End-to-end service analytics covering dependability and performance for big data pipelines, microservices systems and serverless application in IoT-Edge-Cloud environments.
This course is highly designed for PhD and master research level
Agenda:
- Course Overview
- Service-oriented and Cloud Computing: Recap and Outlook
- Lecture: 2h
- Self-study:6h
- Advanced algorithms for Complex and Hybrid cloud systems
- Lecture: 2h
- Self-study and assignment:12h
- Presentation and discussion: 2h
- Coordination and elasticity principles and control algorithms
- Lecture: 2h
- Self-study and assignment:6h
- Complex big data analytics systems and pipelines designs/orchestration
- Lecture: 2h
- Self-study and assignment:12h
- Presentation and discussion: 2h
- End-to-End Service Performance and Dependability Analytics
- Lecture: 2h
- Self-study and assignment:12h
- Presentation and discussion: 2h
- Final exam
- Self-study and preparation: 9h
- Exam: 1h
ECTS-Breakdown:
- 3 ECTS, corresponds to roughly 75 hours of work
- ca. 60% Lecture part, 40% practical part (assignment) = 45h Lecture, 30h Assignments
- Lecture: 11h
- Lecture study: 18h
- Work on assignment topics: 30h
- Presentation and discussion of assignments: 6h
- Exam preparation and exam: 10h
- Total: 45h (Lecture) + 30h (Assignments) = 75h
It is designed for PhD and master research students. It is expected that students are familiar with Internet computing, data analytics, distributed systems, and distributed technologies.
