Dispatching And Scheduling Real-Time Tasks in Multi-Server Systems

Student   Shaik Salman
Advisors   Thomas Nolte
Alessandro V. Papadopoulos
Saad Mubeen
Faculty Reviewer   Marko Bertogna, University of Modena and Reggio Emilia, Modena, Italy
Grading Committee   Dakshina Dasari, Bosch GmbH, Germany
Johan Eker, Lund University, Lund, Sweden
Risat Mahmud Pathan, Chalmers University of Technology, Gothenburg, Sweden
Mats Björkman, Mälardalen University, Sweden (reserve)
Defence   Mälardalen University, Västerås, Sweden
Room Kappa and Zoom meeting (Link will be made public)
November 5th, 2024 13:15
Abstract   Real-time systems such as industrial robots integrate a wide range of algorithms with varying levels of timing requirements to achieve their functional behaviour. Historically, in certain systems, these algorithms were deployed on dedicated single-core hardware platforms that exchanged information over a real-time network while more recent designs have adapted an integrated architecture where these algorithms are executed on an embedded multi-core hardware platform. The advantages provided by cloud and fog architectures for non-real-time applications have prompted discussions around the possibility of achieving similar advantages for systems such as industrial robot controllers by moving from an embedded architecture to a cloud and fog native architecture. Among several challenges that need to be addressed to make such an architecture a practical reality, this thesis focuses on a subset of challenges related to scheduling on multi-server systems. Specifically (1) for the problem of scheduling sequential tasks with clairvoyant inputs and hard-real time requirements, we developed a framework based on minimum parallelism reservations. (2) For the problem of scheduling sequential tasks with stochastic inputs and firm real-time requirements, we developed a framework that utilizes estimated processing times to improve average throughput. (3) Lastly, for the problem of scheduling sequential tasks with arbitrary inputs and soft real-time requirements, we developed competitive algorithms that utilize estimated processing times to minimize average (modified) tardiness in multi-server systems.
Rules and Guidelines   The PhD procedure summary
Rules for Third-cycle Studies at MDH - Chapter 3.1.7 Public Defence of a Thesis
Instructions regarding public defences and licentiate seminars on account of the outbreak of Covid19 (Coronavirus)
Thesis   Thesis
Included Papers   Paper A: Multi-processor scheduling of elastic applications in compositional real-time systems .
Paper B: Evaluating Dispatching and Scheduling Strategies for Firm Real-Time Jobs in Edge Computing .
Paper C: Scheduling firm real-time applications on the edge with single-bit execution time prediction .
Paper D: Dispatching deadline constrained jobs in edge computing systems .
Paper E: Taming Tardiness on Parallel Machines: Online Scheduling with Limited Job Information .
Publications   Complete list of publications

Back to Research

Last modified: 2024-09-13 11:07:08 +0200