Many suspensions, many problems: a review of self-suspending tasks in real-time systems

In general computing systems, a job (process/task) may suspend itself whilst it is waiting for some activity to complete, e.g., an accelerator to return data. In real-time systems, such self-suspension can cause substantial performance/schedulability degradation. This observation, first made in 1988...

Full description

Saved in:
Bibliographic Details
Published in:Real-time systems 2019-01, Vol.55 (1), p.144-207
Main Authors: Chen, Jian-Jia, Nelissen, Geoffrey, Huang, Wen-Hung, Yang, Maolin, Brandenburg, Björn, Bletsas, Konstantinos, Liu, Cong, Richard, Pascal, Ridouard, Frédéric, Audsley, Neil, Rajkumar, Raj, de Niz, Dionisio, von der Brüggen, Georg
Format: Article
Language:English
Subjects:
Algorithms
Communications Engineering
Computer Science
Computer Systems Organization and Communication Networks
Computing time
Control
Mechatronics
Multiprocessing
Networks
Performance and Reliability
Performance degradation
Priority scheduling
Real time
Robotics
Special Purpose and Application-Based Systems
State-of-the-art reviews
Task scheduling
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In general computing systems, a job (process/task) may suspend itself whilst it is waiting for some activity to complete, e.g., an accelerator to return data. In real-time systems, such self-suspension can cause substantial performance/schedulability degradation. This observation, first made in 1988, has led to the investigation of the impact of self-suspension on timing predictability, and many relevant results have been published since. Unfortunately, as it has recently come to light, a number of the existing results are flawed. To provide a correct platform on which future research can be built, this paper reviews the state of the art in the design and analysis of scheduling algorithms and schedulability tests for self-suspending tasks in real-time systems. We provide (1) a systematic description of how self-suspending tasks can be handled in both soft and hard real-time systems; (2) an explanation of the existing misconceptions and their potential remedies; (3) an assessment of the influence of such flawed analyses on partitioned multiprocessor fixed-priority scheduling when tasks synchronize access to shared resources; and (4) a discussion of the computational complexity of analyses for different self-suspension task models.
ISSN:0922-6443
1573-1383
DOI:10.1007/s11241-018-9316-9