Practical Real-Time Virtual Resources: Transparent Task Scheduling on Evenly Distributed Temporal Resource PartitionsSpeaker: Albert M. K. Cheng – Houston, TX, United States
Topic(s): Architecture, Embedded Systems and Electronics, Robotics
Real-time resource partitioning (RP) divides hardware resources (processors, cores, and other components) into temporal partitions and allocates these partitions as virtual resources (physical resources at a fraction of their service rates) to application tasks. RP can be a layer in the OS or firmware directly interfacing the hardware, and is a key enabling technology for virtualization and cloud computing. Open embedded systems make it easy to securely add and remove software applications as well as to increase resource utilization and reduce implementation cost when compared to systems which physically assign distinct computing resources to run different applications. This seminar will describe practical ways based on the Regularity-based Resource Partition Model (RRP) to maintain the schedulability of real-time tasks as if they were scheduled on dedicated physical resources and increase the utilization of the physical multi-resources.
The Hierarchical Real-Time Scheduling (HiRTS) technique helps improve overall resource utilization in real-time embedded systems. With HiRTS, a computation resource is divided into a group of temporal resource partitions, each of which accommodates multiple real-time tasks. Besides the computation resource partitioning problem, real-time task scheduling on resource partitions is also a major problem of HiRTS. The existing scheduling techniques for dedicated resources, like schedulability tests and utilization bounds, are unable to work without changes on temporal resource partitions in most cases. In this seminar, we show how to achieve maximal transparency for task scheduling on Regular Partitions, a type of resource partition introduced by the Regularity-based Resource Partition (RRP) Model. We show that several classes of real-time scheduling problems on a regular partition can be transformed into equivalent problems on a dedicated single resource, such that comprehensive single-resource scheduling techniques provide optimal solutions. Furthermore, this transformation method could be applied to different types of real-time tasks such as periodic tasks, sporadic tasks and aperiodic tasks.
Our research shows that the resource partitioning problem for regular partitions is
much more complicated than the one for periodic partitions. Based on a practical two-layer HiRTS platform, this seminar introduces MulZ (Multiple Z-seqences), which is the first to solve this problem with a partitioned scheduling strategy. By using a more complicated approximation methodology, our experimental results show that MulZ outperforms the current best global scheduling algorithm on this problem. After that, it compares the overall performance of the periodic partition and the regular partition. We conclude that the regular partition is a better choice for the integration of real-time applications.
About this LectureNumber of Slides: 48
Duration: 50 minutes
Languages Available: English, Spanish
Request this Lecture
To request this particular lecture, please complete this online form.
Request a Tour
To request a tour with this speaker, please complete this online form.
All requests will be sent to ACM headquarters for review.