direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Cloud Services

The advent of cloud computing has disruptively changed the way modern applications are developed and operated. Cloud computing is a model that gives cloud users convenient, on-demand network access to a configurable pool of virtualized computing resources offered by a cloud provider. The provisioning and release of capabilities and compute resources occurs rapidly, with minimal management effort and interaction with the provider.

In ISE, we investigate novel cloud technologies, including, but not limited to, new light-weight virtualization approaches for compute resources, e.g., Linux containers, cluster management frameworks, e.g., Kubernetes, Mesos, and new infrastructure management abstractions, e.g., serverless infrastructure provided by services like AWS Lamda or Google Cloud Functions.

Further, we investigate the application of such cutting-edge cloud technologies in different real-world application scenarios, including future energy networks and Internet-of-Things (IoT) scenarios. We identify re-occurring problems in cloud-based application engineering and management and provide proven solutions to address these problems. In particular, we are interested in the quantification of complex cloud service qualities, e.g., consistency, resilience and elasticity, to support engineering and management tasks such as cloud service benchmarking, capacity management, and configuration management.

Microservice architectures provide a conceptual framework for organizations to utilize cloud technologies. In combination with DevOps best practices and workflows such architectures promise to enable IT-centric organizations to improve agility and continuously enhance their business capabilities. The emerging style of "serverless" architectures allows to deliver autonomous applications ("NoOps") and can be viewed as a subform of microservice architecture that speeds up application development and reduces cost of development and operations even more.

Finally, job market trends show an increasing demand for full-stack engineers (full-stack developers), i.e., versatile generalists who are familiar with a range of technologies, such as JavaScript frameworks for cross-platform front-end development (Angular 2, React, React Native, Electron), modern back-end languages and environments (Node.js, Go, Rust), a wide range of distributed system technologies (e.g., NoSQL databases, such as Cassandra, MongoDB, Redis, Riak, HBase, etc.), as well as cloud services and platforms (AWS, Google Cloud Platform, Microsoft Azure). In ISE, we consider cross-stack technology experience to be essential for designing, building, and successfully operating full-stack applications ("you build it, you run it"). Thanks to the abundance of open source web technologies and cloud services, small teams of industrious individuals can quickly develop and operate complex Software-as-a-Service applications.

Related Projects


Anselm Busse and Jacob Eberhard and Sebastian Frost and Dong-Ha Kim and Thore Weißbier and Lukas Renner and Matthias Roth and Stefan Tai (2019). A Response to the United Nations CITES Blockchain Challenge: Incremental and Integrative PoA-based Permit Exchange. 1st IEEE International Conference on Blockchain and Cryptocurrency (ICBC 2019)

Link zur Publikation

D. Ernst and A. Becker and S. Tai (2019). Rapid Canary Assessment Through Proxying and Two-Stage Load Balancing. 4th Workshop on Continuous Software Engineering and 5th Workshop on Quality-Aware DevOps (CSE-QUDOS 2019)

Link zur Publikation

J. Hasenburg and S. Werner and D. Bermbach (2018). Supporting the Evaluation of Fog-based IoT Applications During the Design Phase. Proceedings of the 5th Workshop on Middlware and Applications for the Internet of Things. ACM.

Link zur Publikation

J. Kuhlenkamp and S. Werner (2018). Benchmarking FaaS Platforms: Call for Community Participation. 4th International Workshop on Serverless Computing (WoSC 2018)

Link zur Publikation

S. Werner and J. Kuhlenkamp and M. Klems and Johannes Müller and S. Tai (2018). Serverless Big Data Processing using Matrix Multiplication as Example. IEEE International Conference on BigData 2018

Link zur Publikation

Zusatzinformationen / Extras


Schnellnavigation zur Seite über Nummerneingabe