Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical Results
The concept of a firm real-time task implies the notion of a firm deadline that should not be missed by the jobs of this task. If a deadline miss occurs, the concerned job yields no value to the system. For some applications domains, this restrictive notion can be relaxed. For example, robust contro...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Schloss Dagstuhl -- Leibniz-Zentrum fuer Informatik
2017-02-01
|
| Series: | Leibniz Transactions on Embedded Systems |
| Subjects: | |
| Online Access: | https://drops.dagstuhl.de/storage/07lites/lites_vol004/lites_vol003_issue001/LITES-v004-i001-a002/LITES-v004-i001-a002.pdf |
| _version_ | 1797208380477014016 |
|---|---|
| author | Kluge, Florian |
| author_facet | Kluge, Florian |
| author_sort | Kluge, Florian |
| collection | DOAJ |
| description | The concept of a firm real-time task implies the notion of a firm deadline that should not be missed by the jobs of this task. If a deadline miss occurs, the concerned job yields no value to the system. For some applications domains, this restrictive notion can be relaxed. For example, robust control systems can tolerate that single executions of a control loop miss their deadlines, and still yield an acceptable behaviour. Thus, systems can be developed under more optimistic assumptions, e.g. by allowing overloads. However, care must be taken that deadline misses do not accumulate. This restriction can be expressed by the model of (m,k)-firm real-time tasks that require that from any k consecutive jobs at least m are executed successfully. In this article, we extend our prior work on the MKU scheduling heuristic. MKU uses history-cognisant utility functions as means for making decisions in overload situations. We present new theoretical results on MKU and on other schedulers for (m,k)-firm real-time tasks. Based on extensive simulations, we assess the performance of these schedulers. The results allow us to identify task set characteristics that can be used as guidelines for choosing a scheduler for a concrete use case. |
| first_indexed | 2024-04-24T09:37:53Z |
| format | Article |
| id | doaj.art-a83f131ad22c4b5486b3e3fd7b88743f |
| institution | Directory Open Access Journal |
| issn | 2199-2002 |
| language | English |
| last_indexed | 2024-04-24T09:37:53Z |
| publishDate | 2017-02-01 |
| publisher | Schloss Dagstuhl -- Leibniz-Zentrum fuer Informatik |
| record_format | Article |
| series | Leibniz Transactions on Embedded Systems |
| spelling | doaj.art-a83f131ad22c4b5486b3e3fd7b88743f2024-04-15T07:53:08ZengSchloss Dagstuhl -- Leibniz-Zentrum fuer InformatikLeibniz Transactions on Embedded Systems2199-20022017-02-014102:102:2510.4230/LITES-v004-i001-a002Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical ResultsKluge, Florian0Department of Computer Science, University of AugsburgThe concept of a firm real-time task implies the notion of a firm deadline that should not be missed by the jobs of this task. If a deadline miss occurs, the concerned job yields no value to the system. For some applications domains, this restrictive notion can be relaxed. For example, robust control systems can tolerate that single executions of a control loop miss their deadlines, and still yield an acceptable behaviour. Thus, systems can be developed under more optimistic assumptions, e.g. by allowing overloads. However, care must be taken that deadline misses do not accumulate. This restriction can be expressed by the model of (m,k)-firm real-time tasks that require that from any k consecutive jobs at least m are executed successfully. In this article, we extend our prior work on the MKU scheduling heuristic. MKU uses history-cognisant utility functions as means for making decisions in overload situations. We present new theoretical results on MKU and on other schedulers for (m,k)-firm real-time tasks. Based on extensive simulations, we assess the performance of these schedulers. The results allow us to identify task set characteristics that can be used as guidelines for choosing a scheduler for a concrete use case.https://drops.dagstuhl.de/storage/07lites/lites_vol004/lites_vol003_issue001/LITES-v004-i001-a002/LITES-v004-i001-a002.pdfreal-time scheduling(mk)-firm real-time tasks |
| spellingShingle | Kluge, Florian Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical Results Leibniz Transactions on Embedded Systems real-time scheduling (m k)-firm real-time tasks |
| title | Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical Results |
| title_full | Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical Results |
| title_fullStr | Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical Results |
| title_full_unstemmed | Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical Results |
| title_short | Utility-Based Scheduling of (m,k)-firm Real-Time Tasks - New Empirical Results |
| title_sort | utility based scheduling of m k firm real time tasks new empirical results |
| topic | real-time scheduling (m k)-firm real-time tasks |
| url | https://drops.dagstuhl.de/storage/07lites/lites_vol004/lites_vol003_issue001/LITES-v004-i001-a002/LITES-v004-i001-a002.pdf |
| work_keys_str_mv | AT klugeflorian utilitybasedschedulingofmkfirmrealtimetasksnewempiricalresults |