In heterogeneous environments with performance variations present, multiple applications compete and share a limited amount of system resources, and su#er from variations in resource availability. These complex applications are desired to adapt themselves and to adjust their resource demands dynamically. On one hand, current adaptation mechanisms built within an application cannot preserve global properties such as fairness; on the other hand, adaptive resource management mechanisms built within the operating system are not aware of data semantics in the application. In this paper, we present a novel Middleware Control Framework to enhance the e#ectiveness of QoS adaptation decisions by dynamic control and reconfiguration of internal parameters and functionalities of a distributed multimedia application. Our objective is to satisfy both system-wide properties (such as fairness among concurrent applications) and application-specific requirements (such as preserving the critical performance criteria). The framework is modeled by the Task Control Model and the Fuzzy Control Model, based on rigorous results from the control theory, and verified by the controllability and adaptivity of a distributed visual tracking application. The results show validation of the framework, i.e., critical application quality parameter can be preserved via controlled adaptation.

In this paper, we address power-aware scheduling of periodic hard real-time tasks using dynamic voltage scaling. Our solution includes three parts: (a) a static (off-line) solution to compute the optimal speed, assuming worst-case workload for each arrival, (b) an on-line speed reduction mechanism to reclaim energy by adapting to the actual workload, and (c) an online, adaptive and speculative speed adjustment mechanism to anticipate early completions of future executions by using the average-case workload information. All these solutions still guarantee that all deadlines are met. Our simulation results show that the reclaiming algorithm saves a striking 50% of the energy over the static algorithm. Further, our speculative techniques allow for an additional approximately 20% savings over the reclaiming algorithm. In this study, we also establish that solving an instance of the static power-aware scheduling problem is equivalent to solving an instance of the rewardbased scheduling problem [1, 4] with concave reward functions. 1

Abstract—Reservation and adaptation are two well-known and effective techniques for enhancing the end-to-end performance of network applications. However, both techniques also have limitations, particularly when dealing with high-bandwidth, dynamic flows: fixed-capability reservations tend to be wasteful of resources and hinder graceful degradation in the face of congestion, while adaptive techniques fail when congestion becomes excessive. We propose an approach to quality of service (QoS) that overcomes these difficulties by combining features of reservations and adaptation. In this approach, a combination of online control interfaces for resource management, a sensor permitting online monitoring, and decision procedures embedded in resources enable a rich variety of dynamic feedback interactions between applications and resources. We describe a QoS architecture, GARA, that has been extended to support these mechanisms, and use three examples of application-level adaptive strategies to show how this framework can permit applications to adapt both their resource requests and behavior in response to online sensor information. I.

In this paper, we consider the use of an application-layer performance measure --- the utility --- in the context of bandwidth allocation for an available bit rate service. Our bandwidth allocation scheme can be viewed as a generalization of traditional available bit rate service; our scheme is equivalent to bandwidth max-min allocation when the utility of all applications are equal. The goal of our allocation scheme is to provide good application-layer service to a wide diversity of applications sharing available bandwidth. We achieve this goal while also supporting changes in utility over time, tolerating some inaccuracy in utility function specification, and addressing the issue of circumvention through pricing. Keywords---Utility functions, available bit rate. I. INTRODUCTION A N important class of distributed applications are those that can adapt their resource usage based on feedback from the network. These applications are robust against variability in network performance an...

In a distributed environment where multiple applications compete and share a limited amount of system resources, applications tend to suffer from variations in resource availability, and are desired to adapt their behavior to the resource variations of the system. We propose a Task Control Model to rigorously model the dynamics of an adaptive system, using the digital control theory. With our Task Control Model, we are able to quantitatively analyze the stability and equilibrium of the adaptive applications, while simultaneously providing fairness guarantees to other applications in the system. Our control algorithm has also been extended to the cases where no sufficient task state information are observable. We show that even under these circumstances, our Task Control Model can still be applied and our control algorithms yield stable and responsive behavior. 1 Introduction In a heterogeneous distributed environment, a wide variety of distributed applications demand specific Quality ...

Reward-based scheduling refers to the problem in which there is a reward associated with the execution of a task. In our framework, each real-time task comprises a mandatory and an optional part, with which a nondecreasing reward function is associated. Imprecise computation and Increased-Reward-with-Increased-Service models fall within the scope of this framework. In this paper, we address the reward-based scheduling problem for periodic tasks. For linear and concave reward functions we show: (a) the existence of an optimal schedule where the optional service time of a task is constant at every instance and (b) how to efficiently compute this service time. We also prove that RMS (with harmonic periods), EDF and LLF policies are optimal when used with the optimal service times we computed, and that the problem becomes NP-Hard, when the reward functions are convex. Further, our solution eliminates runtime overhead, and makes possible the use of existing scheduling disciplines.

Memory can be efficiently utilized if the dynamic memory demands of applications can be determined and analyzed at run-time. The page miss ratio curve(MRC), i.e. page miss rate vs. memory size curve, is a good performance-directed metric to serve this purpose. However, dynamically tracking MRC at run time is challenging in systems with virtual memory because not every memory reference passes through the operating system (OS). This paper

In this paper we present a framework for dynamically allocating the CPU resource to tasks whose execution times are not known a priori. Tasks are partitioned in three classes: the ones that require a uniform execution but do not impose any temporal constraint, periodic tasks that operate on Continuous Media, and event driven tasks that respond to external interrupts. For the last two classes, we show how to adjust the fraction of the CPU bandwidth assigned to each task using a feedback mechanism.

The paper presents a feedback scheduling mechanism in the context of co-design of the scheduler and the control tasks. We are particularly interested in controllers where the execution time may change abruptly between different modes, such as in hybrid controllers. The proposed solution attempts to keep the CPU utilization at a high level, avoid overload, and distribute the computing resources evenly among the tasks. The feedback scheduler is implemented as a periodic or sporadic task that assigns sampling periods to the controllers based on execution-time measurements. The controllers may also communicate feedforward mode-change information to the scheduler. As an example, we consider hybrid control of a set of double-tank processes. The system is evaluated, from both scheduling and control performance perspectives, by co-simulation of controllers, scheduler, and tanks. 1.

... this paper we propose a solution to this problem; to our knowledge, this is the first solution that combines the three constraints mentioned above. We devise two algorithms, an optimal algorithm for homogeneous applications (with respect to power consumption) and a heuristic iterative algorithm that can also accommodate heterogeneous applications (that is, those with different power consumption functions). We show by simulation that our iterative algorithm is fast and within 1% of the optimal.