Methods for power optimization in distributed embedded systems with real-time requirements

Racu, Razvan; Hamann, Arne; Ernst, Rolf; Mochocki, B.; Hu, Xiaobo Sharon

doi:10.1145/1176760.1176806

Cited by 8 publications

(4 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Applications or tasks with variable, different behaviours, and platforms that can adapt/reconfigure are considered. For example, [30] proposes a method to reduce energy consumption in a system executing a mix of periodic, aperiodic, and bursty tasks. [36] targets a system with a unified support for hardware, and operating point configuration methods.…”

Section: Run-time Energy Managementmentioning

confidence: 99%

A composable, energy-managed, real-time MPSOC platform

Molnos

Ambrose

Nelson

et al. 2010

2010 12th International Conference on Optimization of Electrical and Electronic Equipment

View full text Add to dashboard Cite

Abstract-Multi-processors systems on chip (MPSOC) platforms emerged in embedded systems as hardware solutions to support the continuously increasing functionality and performance demands in this domain. Such a platform has to execute a mix of applications with diverse performance and timing constraints, i.e., real-time or non-real-time, thus different application schedulers should co-exist on an MPSOC. Moreover, applications share many MPSOC resources, thus their timing depends on the arbitration at these resources. Arbitration may create inter-application dependencies, e.g., the timing of a low priority application depends on the timing of all higher priority ones. Application inter-dependencies make the functional and timing verification and the integration process harder. This is especially problematic for real-time applications, for which fulfilling the time-related constraints should be guaranteed by construction. Moreover, energy and power management, commonly employed in embedded systems, make this verification even more difficult. Typically, energy and power management involves scaling the resources operating point, which has a direct impact on the resource performance, thus influences the application time behaviour. Finally, a small change in one application leads to the need to re-verify all other applications, incurring a large effort. Composability is a property meant to ease the verification and integration process. A system is composable if the functionality and the timing behaviour of each application is independent of other applications mapped on the same platform. Composability is achieved by utilising arbiters that ensure applications independence. In this paper we present the concepts behind a composable, scalable, energy-managed MPSOC platform, able to support different real-time and nonreal time schedulers concurrently, and discuss its advantages and limitations.

show abstract

Section: Run-time Energy Managementmentioning

confidence: 99%

A composable, energy-managed, real-time MPSOC platform

Molnos

Ambrose

Nelson

et al. 2010

2010 12th International Conference on Optimization of Electrical and Electronic Equipment

View full text Add to dashboard Cite

show abstract

“…Works addressing voltage scaling for hard real-time task sets are [1] and [2], [3], and [4], whereof the former two works consider periodic task sets only, and the latter two are able to optimise for other occurrence behaviour as well. Nevertheless, all four works neglect transition costs for mode switching.…”

Section: Dynamic Voltage Scalingmentioning

confidence: 99%

Task-Dependent Processor Shutdown for Hard Real-Time Systems

Lipskoch

Slomka²

2009

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Abstract. Mobile devices relying on batteries can save energy by using lowpower modes of their processors. In a hard real-time environment, one has to prove the real-time feasibility and thereby to guarantee that energy saving methods do not violate real-time constraints. Besides the processor's unavailability during low-power mode, the transition to and from the mode consumes energy and time.This work introduces a task-dependent policy for mode switching and compares it to procrastination techniques from the literature. The low-power interval is placed between occurrences of one task of the task set such that low-power mode and instances of this task do not overlap.Optimisation of task to depend on, duration of low-power, and shutdown rate is done with the help of a hard real-time test to provide feasible results. The used test provides analysis for preemptible, deadline scheduled task sets. Tasks are allowed to have periodic, periodic with jitter, sporadic, or other behaviour regarding their occurrence.Thus, this work extends the applicability of processor shutdown to such environments. And since parameters are determined off-line, apart form a programmable real-time clock, no power consuming extra circuitry is needed. However, the method comes with a slight modification of the task set.

show abstract

“…Some well-established analytical methods and tools potentially interacting with complex analysis scenarios include design space exploration and sensitive analysis [12] [13].…”

Section: Complex Evaluation Scenariosmentioning

confidence: 99%

“…It allows the system designer to assess the system-level impact of changes in nonfunctional properties of individual hardware and software components. Particularly for scheduling-aware analysis techniques, there is a number of works providing search algorithms to onedimensional and multidimensional (simultaneous variations of a set of non-functional properties) sensitivity analysis [13]. Although the sensitivity analysis process may be fully delegated to analysis tools, there is an important gap in MDE for addressing the correct parameterization (e.g., determination of dependent, independent, and critical parameters) and solution selection from a modeling viewpoint:…”

Section: Issue 2: "How Can We Express the Relationships Between Diffementioning

confidence: 99%

Leveraging analysis-aided design decision knowledge in UML-based development of embedded systems

Espinoza

Servat

Gérard

2008

Proceedings of the 3rd International Workshop on Sharing and Reusing Architectural Knowledge

View full text Add to dashboard Cite

Many important works have been carried out to provide modeling languages (e.g., UML, SDL) with expressiveness to support embedded system design, validation and verification. A fundamental shortcoming in current model-driven approaches is the inability to explicitly capture design decisions and trade-offs between different non-functional parameters, among which timeliness, memory usage, and power consumption are of primary interest. This paper highlights technical limitations in UML to specify complex non-functional evaluation scenarios of candidate architectures, and outlines our current work to provide straightforward solutions.

show abstract

Methods for power optimization in distributed embedded systems with real-time requirements

Cited by 8 publications

References 40 publications

A composable, energy-managed, real-time MPSOC platform

A composable, energy-managed, real-time MPSOC platform

Task-Dependent Processor Shutdown for Hard Real-Time Systems

Leveraging analysis-aided design decision knowledge in UML-based development of embedded systems

Contact Info

Product

Resources

About