direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

MA: Embedding IoT Devices into Distributed Dataflow Frameworks

The Internet of Things (IoT) is characterized by an increasing number of small devices connected to the global internet. With applications like connected cars or the monitoring of emissions and ambient pollution, IoT gradually becomes part of the everyday life. Typically, the data generated by IoT device must be processed in some way to extract valuable information. Given the number of IoT devices and the global scale of their deployments, data processing must be performed in a distributed manner.

A tried and tested way to develop distributed applications is the use of distributed dataflow frameworks like Apache Spark or Apache Flink. Those frameworks let users write their distributed applications by means of sequential building blocks that, then, are connected to a directed acyclic graph (DAG) which forms the dataflow. The frameworks not only take care of distributing and parallelizing the applications but also provide features like fault tolerance.

However, while the dataflow frameworks help users develop distributed applications, they typically assume an underlying cluster of commodity nodes in a local network. As such, resource-constrained IoT devices that are globally distributed are treated as second-class citizens.

The aim of this bachelor’s/master’s thesis is to tackle this exact issue and enable a more seamless integration of IoT devices into distributed dataflow frameworks. Theses in this area include improving the scheduling of dataflow frameworks in order to factor in the way how IoT devices are geographically distributed or mechanisms that enable dataflow operators to be scheduled and executed on such devices.


Ilya Verbitskiy
+49 30 314-26978
Room TEL 1209
e-mail query [1]
------ Links: ------

Zusatzinformationen / Extras

Quick Access:

Schnellnavigation zur Seite über Nummerneingabe

Auxiliary Functions

Copyright TU Berlin 2008