Results 1 
7 of
7
EnergyEfficient, Utility Accrual RealTime Scheduling Under the Unimodal Arbitrary Arrival Model
 in ACM Design, Automation, and Test in Europe (DATE
, 2005
"... We present an energyefficient realtime scheduling algorithm called EUA∗, for the unimodal arbitrary arrival model (or UAM). UAM embodies a “stronger ” adversary than most arrival models. The algorithm considers application activities that are subject to time/utility function time constraints, UAM, ..."
Abstract

Cited by 22 (5 self)
 Add to MetaCart
(Show Context)
We present an energyefficient realtime scheduling algorithm called EUA∗, for the unimodal arbitrary arrival model (or UAM). UAM embodies a “stronger ” adversary than most arrival models. The algorithm considers application activities that are subject to time/utility function time constraints, UAM, and the multicriteria scheduling objective of probabilistically satisfying utility lower bounds, and maximizing systemlevel energy efficiency. Since the scheduling problem is intractable, EUA ∗ allocates CPU cycles, scales clock frequency, and heuristically computes schedules using statistical estimates of cycle demands, in polynomialtime. We establish that EUA ∗ achieves optimal timeliness during underloads, and identify the conditions under which timeliness assurances hold. Our simulation experiments illustrate EUA∗’s superiority. 1.
A StandbySparing Technique with Low EnergyOverhead for FaultTolerant Hard RealTime Systems
 in Proc. 7th Int’l Conf. Hardware/Software Codesign and Sys. Synthesis (CODES+ISSS’09
, 2009
"... Time redundancy (rollbackrecovery) and hardware redundancy are commonly used in realtime systems to achieve fault tolerance. From an energy consumption point of view, time redundancy is generally more preferable than hardware redundancy. However, hard realtime systems often use hardware redundanc ..."
Abstract

Cited by 12 (1 self)
 Add to MetaCart
(Show Context)
Time redundancy (rollbackrecovery) and hardware redundancy are commonly used in realtime systems to achieve fault tolerance. From an energy consumption point of view, time redundancy is generally more preferable than hardware redundancy. However, hard realtime systems often use hardware redundancy to meet high reliability requirements of safetycritical applications. In this paper we propose a hardwareredundancy technique with low energyoverhead for hard realtime systems. The proposed technique is based on standbysparing, where the system is composed of a primary unit and a spare. Through analytical models, we have developed an online energymanagement method which uses a slack reclamation scheme to reduce the energy consumption of both the primary and spare units. In this method, dynamic voltage scaling (DVS) is used for the primary unit and dynamic power management (DPM) is used for the spare. We conducted several experiments to compare the proposed system with a faulttolerant realtime system which uses time redundancy for fault tolerance and DVS with slack reclamation for low energy consumption. The results show that for relaxed time constraints, the proposed system provides up to 24 % energy saving as compared to the timeredundancy system. For tight deadlines when the timeredundancy system can tolerate no faults, the proposed system preserves its faulttolerance but with about 32% more energy consumption.
Utility Accrual RealTime Scheduling Under Variable Cost Functions
, 2005
"... We present a utility accrual realtime scheduling algorithm called CICVCUA, for tasks whose execution times are functions of their starting times. We model such variable execution times employing variable cost functions (or VCFs). The algorithm considers application activities that are subject to t ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
(Show Context)
We present a utility accrual realtime scheduling algorithm called CICVCUA, for tasks whose execution times are functions of their starting times. We model such variable execution times employing variable cost functions (or VCFs). The algorithm considers application activities that are subject to time/utility function time constraints (or TUFs), execution times described using VCFs, and concurrent, mutually exclusive sharing of nonCPU resources. We consider the multicriteria scheduling objective of (1) assuring that the maximum interval between any two consecutive, successful completions of jobs of a task must not exceed a specified upper bound, and (2) maximizing the system’s total accrued utility, while satisfying mutual exclusion resource constraints. Since the scheduling problem is intractable, CICVCUA statically computes worstcase sojourn times of tasks, selects tasks for execution based on their potential utility density, and completes them at specific times, in polynomialtime. We establish that CICVCUA achieves optimal timeliness during underloads. Further, we identify the conditions under which timeliness assurances hold. Our simulation experiments illustrate CICVCUA’s effectiveness and superiority. Acknowledgments
LowEnergy StandbySparing for Hard RealTime Systems
"... Abstract — Timeredundancy techniques are commonly used in realtime systems to achieve fault tolerance without incurring high energy overhead. However, reliability requirements of hard realtime systems that are used in safetycritical applications are so stringent that timeredundancy techniques a ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
(Show Context)
Abstract — Timeredundancy techniques are commonly used in realtime systems to achieve fault tolerance without incurring high energy overhead. However, reliability requirements of hard realtime systems that are used in safetycritical applications are so stringent that timeredundancy techniques are sometimes unable to achieve them. Standby sparing as a hardwareredundancy technique can be used to meet high reliability requirements of safetycritical applications. However, conventional standbysparing techniques are not suitable for lowenergy hard realtime systems as they either impose considerable energy overheads or are not proper for hard timing constraints. In this paper we provide a technique to use standby sparing for hard realtime systems with limited energy budgets. The principal contribution of this work is an online energymanagement technique which is specifically developed for standbysparing systems that are used in hard realtime applications. This technique operates at runtime and exploits dynamic slacks to reduce the energy consumption while guaranteeing hard deadlines. We compared the lowenergy standbysparing (LESS) system with a lowenergy timeredundancy system (from a previous work). The results show that for relaxed time constraints, the LESS system is more reliable and provides about 26 % energy saving as compared to the timeredundancy system. For tight deadlines when the timeredundancy system is not sufficiently reliable (for safetycritical application), the LESS system preserves its reliability but with about 49 % more energy consumption.
iii Contents List of Tables.................................. vi
, 2004
"... Worstcase time predictions for soft realtime robotic systems by Johannes Stickel, ..."
Abstract
 Add to MetaCart
(Show Context)
Worstcase time predictions for soft realtime robotic systems by Johannes Stickel,
Technical Report No. 2005499 Scheduling Algorithms for RealTime Systems ∗
, 2005
"... The problem of realtime scheduling spans a broad spectrum of algorithms from simple uniprocessor to highly sophisticated multiprocessor scheduling algorithms. In this paper, we study the characteristics and constraints of realtime tasks which should be scheduled to be executed. Analysis methods an ..."
Abstract
 Add to MetaCart
(Show Context)
The problem of realtime scheduling spans a broad spectrum of algorithms from simple uniprocessor to highly sophisticated multiprocessor scheduling algorithms. In this paper, we study the characteristics and constraints of realtime tasks which should be scheduled to be executed. Analysis methods and the concept of optimality criteria, which leads to design appropriate scheduling algorithms, will also be addressed. Then, we study realtime scheduling algorithms for uniprocessor systems, which can be divided into two major classes: offline and online. Online algorithms are partitioned into either static or dynamicpriority based algorithms. We discuss both preemptive and nonpreemptive staticpriority based algorithms. For dynamicpriority based algorithms, we study the two subsets; namely, planning based and best effort scheduling algorithms. Some of the uniprocessor scheduling algorithms are illustrated by examples in the Appendix. Multiprocessor scheduling algorithms is another class of realtime scheduling algorithms which is discussed in the paper as well. We also describe techniques to deal with aperiodic and sporadic tasks, precedence constraints, and priority inversion.
AN APPROACH TO REAL TIME ADAPTIVE DECISION MAKING IN DYNAMIC DISTRIBUTED SYSTEMS
, 2005
"... Efficient operation of a dynamic system requires (near) optimal realtime control decisions. Those decisions depend on a set of control parameters that change over time. Very often, the optimal decision can be made only with the knowledge of future values of control parameters. As a consequence, the ..."
Abstract
 Add to MetaCart
(Show Context)
Efficient operation of a dynamic system requires (near) optimal realtime control decisions. Those decisions depend on a set of control parameters that change over time. Very often, the optimal decision can be made only with the knowledge of future values of control parameters. As a consequence, the decision process is heuristic in nature. The optimal decision can be determined only after the fact, once the uncertainty is removed. For some types of dynamic systems, the heuristic approach can be very effective. The basic premise is that the future values of control parameters can be predicted with sufficient accuracy. We can either predict those value based on a good model of the system or based on historical data. In many cases, the good model is not available. In that case, prediction using historical data is the only option. It is necessary to detect similarities with the current situation and extrapolate future values. In other words, we need to (quickly) identify patterns in historical data that match the current data pattern. The low sensitivity of the optimal solution is critical. Small variations in data patterns should affect minimally the optimal solution. Resource allocation problems and other “discrete decision systems ” are good examples of such systems. The main contribution of this work is a novel heuristic methodology that uses neural