Tasklets: an approach to best-effort computing

• Applicant: Professor Dr. Christian Becker
• Subject Area: Security and Dependability, Operating-, Communication- and Distributed Systems
• Term: from 2016 to 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 323412509

• Final Report Year: 2023

Final Report Abstract

The overall goal of this project was the investigation of a light weight computation concept that should allow distributed computation in heterogeneous environments. Computer systems these days range from small, embedded systems, mobile device, multi- and many-core processors and powerful GPUs. These computing systems are connected by networks of varying characteristics w.r.t. delay, jitter and availability. Thus, interrupted or delayed computation is not an error but a characteristic of such systems. By providing a light weight virtual machine computation becomes homogeneous and available on a large range of platforms. In order to allow for compensation of connection losses, termination of cloud instances or other irregularities in execution, the concept of Quality of Computation (QoC) is introduced and handled by an orchestration service. Thus applications can specify their requirements on QoC and the orchestration service is responsible to find candidates for execution.

Publications

• Tasklets: "Better than Best-Effort" Computing. 2016 25th International Conference on Computer Communication and Networks (ICCCN), 1-11. IEEE.
  Schafer, Dominik; Edinger, Janick; Paluska, Justin Mazzola; VanSyckel, Sebastian & Becker, Christian
  
• Fault-avoidance strategies for context-aware schedulers in pervasive computing systems. 2017 IEEE International Conference on Pervasive Computing and Communications (PerCom), 79-88. IEEE.
  Edinger, Janick; Schafer, Dominik; Krupitzer, Christian; Raychoudhury, Vaskar & Becker, Christian
  
• Hybrid Task Scheduling for Mobile Devices in Edge and Cloud Environments. 2018 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 669-674. IEEE.
  Schafer, Dominik; Edinger, Janick; Eckrich, Jens; Breitbach, Martin & Becker, Christian
  
• Self-adaptive Resource Allocation for Continuous Task Offloading in Pervasive Computing. 2018 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), 663-668. IEEE.
  Choochotkaew, Sunyanan; Yamaguchi, Hirozumi; Higashino, Teruo; Schafer, Dominik; Edinger, Janick & Becker, Christian
  
• Workload Partitioning and Task Migration to Reduce Response Times in Heterogeneous Computing Environments. 2018 27th International Conference on Computer Communication and Networks (ICCCN), 1-11. IEEE.
  Schafer, Dominik; Edinger, Janick; Breitbach, Martin & Becker, Christian
  
• Context-Aware Data and Task Placement in Edge Computing Environments. 2019 IEEE International Conference on Pervasive Computing and Communications (PerCom, 1-10. IEEE.
  Breitbach, Martin; Schafer, Dominik; Edinger, Janick & Becker, Christian
  
• DataVinci: Proactive Data Placement for Ad-Hoc Computing. 2021 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), 866-873. IEEE.
  Breitbach, Martin; Edinger, Janick; Schafer, Dominik & Becker, Christian
  
• Decentralized Low-Latency Task Scheduling for Ad-Hoc Computing. 2021 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 776-785. IEEE.
  Edinger, Janick; Breitbach, Martin; Gabrisch, Niklas; Schafer, Dominik; Becker, Christian & Rizk, Amr
  
• Voltaire: Precise Energy-Aware Code Offloading Decisions with Machine Learning. 2021 IEEE International Conference on Pervasive Computing and Communications (PerCom), 1-10. IEEE.
  Breitbach, Martin; Edinger, Janick; Kaupmees, Siim; Trotsch, Heiko; Krupitzer, Christian & Becker, Christian