March 3 – 8 , 2019, Dagstuhl Seminar 19101

Analysis, Design, and Control of Predictable Interconnected Systems


Kunal Agrawal (Washington University – St. Louis, US)
Enrico Bini (University of Turin, IT)
Jens Schmitt (TU Kaiserslautern, DE)
Giovanni Stea (University of Pisa, IT)

For support, please contact

Dagstuhl Service Team


Dagstuhl Report, Volume 9, Issue 3 Dagstuhl Report
Aims & Scope
List of Participants
Dagstuhl's Impact: Documents available


Computing capacity surronds our environment more and more. The exploitation of this diffused infrastructure poses a number of challenges. A notable one is the guarantee of predictable performance. Many modern computing platforms require high degrees of predictability in their timing characteristics -- for instance, in avionics and automotive applications, it is essential that the tasks complete in a timely manner in order to take appropriate actions in response to a developing situation. With increasing complexity and heterogeneity in functionality required by time-sensitive applications, there is an increasing need for designing distributed interconnected platforms that respond to time-varying requests in a highly predictable way.

The world of networking is undergoing an analogous transformation. There is a major shift towards smarter and more autonomous networks, the so-called self-driving networks. The goal is to mimic the success that cloud-computing techniques and concepts had on the transformation of the IT infrastructure. The latter made it possible to allow physical resources, such as computing and storage nodes, to be shared among users through the use virtual resources (Network Function Virtualization). Virtualization of network functions offers an efficient way to meet dynamic user demands in a cost-effective manner. However, the guarantee of predictability in NFV is, to date, an open problem.

With increasing complexity and heterogeneity in functionality required by timing-sensitive applications, there is need for designing distributed and interconnected platforms that provide high predictability. On the one hand, with the advent of Cyber-physical systems and Industry 4.0, real-time systems are becoming more and more networked systems. On the other hand, networking scenarios where a-priori quantifiable guarantees on the worst-case traversal time are required as a prerequisite for a correct application-level computation are becoming commonplace. Therefore, it is necessary to develop methods to ensure compliance with end-to-end deadlines in distributed systems where both computation and communication are involved. Dagstuhl seminar "Analysis, design and control of predictable interconnected systems" gathered more than 40 researchers from four continents, with a good balance of seniority and gender. These researchers came from different communities, including Network Calculus, Real-time Systems, Control Theory, Performance Evaluation and Data-flow, which have been discussing similar problems in the recent past, sometimes also using similar methodologies, but different notations and hypotheses, and focusing on different characteristics. The aim of the seminar was to foster cross-fertilization and inter-community networking, using new, contemporary problems which are interesting for and discussed within more than one community. The seminar hosted 26 talks. Some of these were "background-levelling" talks given by well-recognized senior experts in the resepective communities, with the aim of crossing the lexicon and notation gap between communities. Other talks, building on the shared background provided by the former, discussed interesting novel problems and promising application areas. The open, schedule-as-you-go format for the talks and the time left open for free networking activities fostered an open environment. The general conclusion, which can be gathered by the (mostly enthusiastic) comments in the inal survey, is that the involved communities have more in common with each other than the attendees thought, and cross-fertilization is necessary to tackle new problems in a holistic approach.

Summary text license
  Creative Commons BY 3.0 Unported license
  Kunal Agrawal, Enrico Bini, and Giovanni Stea


  • Networks
  • Optimization / Scheduling


  • Real-time systems
  • Network calculus
  • Distributed resource management


In the series Dagstuhl Reports each Dagstuhl Seminar and Dagstuhl Perspectives Workshop is documented. The seminar organizers, in cooperation with the collector, prepare a report that includes contributions from the participants' talks together with a summary of the seminar.


Download overview leaflet (PDF).

Dagstuhl's Impact

Please inform us when a publication was published as a result from your seminar. These publications are listed in the category Dagstuhl's Impact and are presented on a special shelf on the ground floor of the library.


Furthermore, a comprehensive peer-reviewed collection of research papers can be published in the series Dagstuhl Follow-Ups.