Jianguo Yao

Online adaptive utilization control for real-time embedded multiprocessor systems

To provide Quality of Service (QoS) guarantees in open and unpredictable environments, the utilization control problem is defined to keep the processor utilization at the schedulable utilization bound, even in the face of unpredictable and/or varying task execution times. To handle the end-to-end task model where each task is comprised of a chain of subtasks distributed on multiprocessors, researchers have used Model Predictive Control (MPC) to address the Multiple-Input, Multiple-Output (MIMO) control problem. Although MPC can handle a limited range of model uncertainties due to execution time estimation errors, the system may suffer performance deterioration or even become unstable if the actual task execution times are much larger than their estimated values. In this paper, we present an online adaptive optimal control approach using Recursive Least Squares (RLS) based model estimator plus Linear Quadratic (LQ) optimal controller. We use simulation experiments to demonstrate the effectiveness of our controller compared with the MPC-based controller.

Adaptive Calibration for Fusion-based Wireless Sensor Networks

Wireless sensor networks (WSNs) are typically composed of low-cost sensors that are deeply integrated with physical environments. As a result, the sensing performance of a WSN is inevitably undermined by various physical uncertainties, which include stochastic sensor noises, unpredictable environment changes and dynamics of the monitored phenomenon. Traditional solutions (e.g., sensor calibration and collaborative signal processing) work in an open-loop fashion and hence fail to adapt to these uncertainties after system deployment. In this paper, we propose an adaptive system-level calibration approach for a class of sensor networks that employ data fusion to improve system sensing performance. Our approach features a feedback control loop that exploits sensor heterogeneity to deal with the aforementioned uncertainties in calibrating system performance. In contrast to existing heuristic based solutions, our control-theoretical calibration algorithm can ensure provable system stability and convergence. We also systematically analyze the impacts of communication reliability and delay, and propose an optimal routing algorithm that minimizes the impact of packet loss on system stability. Our approach is evaluated by both experiments on a testbed of Tmotes as well as extensive simulations based on data traces gathered from a real vehicle detection experiment. The results demonstrate that our calibration algorithm enables a network to maintain the optimal detection performance in the presence of various system and environmental dynamics.

Dynamic Control of Electricity Cost with Power Demand Smoothing and Peak Shaving for Distributed Internet Data Centers

Internet based service providers, such as Amazon, Google, Yahoo etc, build their data centers (IDC) across multiple regions to provide reliable and low latency of services to clients. Ever-increasing service demand, complexity of services and growing client population cause enormous power consumptions by these IDCs incurring a major part of their running costs. Modern electric power grid provides a feasible way to dynamically and efficiently manage the electricity cost of distributed IDCs based on the Locational Marginal Pricing (LMP) policy. While recent works exploit LMP by electricity-price based geographic load distribution, the dynamic workload and high volatility of electricity prices induce highly volatile power demand and critical power peak problem. The benefit of cost minimization via geographic load distribution is counterbalanced with the high cost incurred by violating the peak power. In this paper, we study the dynamic control of electricity cost to provide low volatility in power demand and shaving of power peaks. To this end, a Model Predictive Control (MPC) electricity cost minimization problem is formulated based on a time-continuous differential model. The proposed solution minimizes electricity costs, provides low variation in power demand by penalizing the change in workload and alleviates the power peaks by tracking the available power budget. By providing extensive simulation results based on real-life electricity price traces we show the effectiveness of our approach.

NetSimplex: Controller Fault Tolerance Architecture in Networked Control Systems

The assurance of reliability becomes increasingly challenging as the complexity of networked control systems (NCS) rapidly increases. Simplex architecture was designed to tolerate control software design and implementation. This architecture consists of a high assurance controller (HAC) and a high performance controller (HPC). The HAC uses the linear state feedback control to create a large maximum stability region (MSR). The HPC aims at achieving a better control performance and may use any design. However, the plants states under HPC must stay within the MSR, or the control is switched to HAC.

vGASA: Adaptive Scheduling Algorithm of Virtualized GPU Resource in Cloud Gaming

As the virtualization technology for GPUs matures, cloud gaming has become an emerging application among cloud services. In addition to the poor default mechanisms of GPU resource sharing, the performance of cloud games is inevitably undermined by various runtime uncertainties such as rendering complex game scenarios. The question of how to handle the runtime uncertainties for GPU resource sharing remains unanswered. To address this challenge, we propose vGASA, a virtualized GPU resource adaptive scheduling algorithm in cloud gaming. vGASA interposes scheduling algorithms in the graphics API of the operating system, and hence the host graphic driver or the guest operating system remains unmodified. To fulfill the service level agreement as well as maximize GPU usage, we propose three adaptive scheduling algorithms featuring feedback control that mitigates the impact of the runtime uncertainties on the system performance. The experimental results demonstrate that vGASA is able to maintain frames per second of various workloads at the desired level with the performance overhead limited to 5-12 percent.

Differential privacy in telco big data platform

Differential privacy (DP) has been widely explored in academia recently but less so in industry possibly due to its strong privacy guarantee. This paper makes the first attempt to implement three basic DP architectures in the deployed telecommunication (telco) big data platform for data mining applications. We find that all DP architectures have less than 5% loss of prediction accuracy when the weak privacy guarantee is adopted (e.g., privacy budget parameter e ≥ 3). However, when the strong privacy guarantee is assumed (e.g., privacy budget parameter e ≤ 0:1), all DP architectures lead to 15% ~ 30% accuracy loss, which implies that real-word industrial data mining systems cannot work well under such a strong privacy guarantee recommended by previous research works. Among the three basic DP architectures, the Hybridized DM (Data Mining) and DB (Database) architecture performs the best because of its complicated privacy protection design for the specific data mining algorithm. Through extensive experiments on big data, we also observe that the accuracy loss increases by increasing the variety of features, but decreases by increasing the volume of training data. Therefore, to make DP practically usable in large-scale industrial systems, our observations suggest that we may explore three possible research directions in future: (1) Relaxing the privacy guarantee (e.g., increasing privacy budget e) and studying its effectiveness on specific industrial applications; (2) Designing specific privacy scheme for specific data mining algorithms; and (3) Using large volume of data but with low variety for training the classification models.

VGRIS: virtualized GPU resource isolation and scheduling in cloud gaming

Fueled by the maturity of virtualization technology for Graphics Processing Unit (GPU), there is an increasing number of data centers dedicated to GPU-related computation tasks in cloud gaming. However, GPU resource sharing in these applications is usually poor. This stems from the fact that the typical cloud gaming service providers often allocate one GPU exclusively for one game. To achieve the efficiency of computational resource management, there is a demand for cloud computing to employ the multi-task scheduling technologies to improve the utilization of GPU. In this paper, we propose VGRIS, a resource management framework for Virtualized GPU Resource Isolation and Scheduling in cloud gaming. By leveraging the mature GPU paravirtualization architecture, VGRIS resides in the host through library API interception, while the guest OS and the GPU computing applications remain unmodified. In the proposed framework, we implemented three scheduling algorithms in VGRIS for different objectives, i.e., Service Level Agreement (SLA)-aware scheduling, proportional-share scheduling, and hybrid scheduling that mixes the former two. By designing such a scheduling framework, it is possible to handle different kinds of GPU computation tasks for different purposes in cloud gaming. Our experimental results show that each scheduling algorithm can achieve its goals under various workloads.

Predictive Electricity Cost Minimization Through Energy Buffering in Data Centers

More and more cloud computing services are handled by different Internet operators in distributed Internet data centers (IDCs), which incurs massive electricity costs. Today, the power usage of data centers contributes to more than 1.5% market share of electricity consumption across the United States. Minimization of these costs benefits cloud computing operators, and attracts increasing attentions from many research groups and industrial sectors. Along with the deployment of smart grid, the electrical real-time pricing policy promotes power consumers to adaptively schedule their electricity utilization for lower operational costs. This paper proposes a novel approach to enable electrical energy buffering in batteries to predictively minimize IDC electricity costs in smart grid. Batteries are charged when electricity price is low and discharged to power servers when electricity price is high. A power management controller is used per battery to arbitrate the charging and discharging actions of the battery. The controller is designed as a MPC-based (model predictive control) controller. To this end, an MPC power minimization problem is formulated based on a discrete state-space model with states of battery power level and cost. Extensive simulation results demonstrate the effectiveness of our approach based on real-life electricity prices in smart grid.

Power Admission Control With Predictive Thermal Management in Smart Buildings

This paper presents a control scheme for thermal management in smart buildings based on predictive power admission control. This approach combines model predictive control with budget-schedulability analysis in order to reduce peak power consumption as well as ensure thermal comfort. First, the power budget with a given thermal comfort constraint is optimized through budget-schedulability analysis which amounts to solving a constrained linear programming problem. Second, the effective peak power demand is reduced by means of the optimal scheduling and cooperative operation of multiple thermal appliances. The performance of the proposed control scheme is assessed by simulation based on the thermal dynamics of a real eight-room office building located at Danish Technical University.

VGRIS: Virtualized GPU Resource Isolation and Scheduling in Cloud Gaming

To achieve efficient resource management on a graphics processing unit (GPU), there is a demand to develop a framework for scheduling virtualized resources in cloud gaming. In this article, we propose VGRIS, a resource management framework for virtualized GPU resource isolation and scheduling in cloud gaming. A set of application programming interfaces (APIs) is provided so that a variety of scheduling algorithms can be implemented within the framework without modifying the framework itself. Three scheduling algorithms are implemented by the APIs within VGRIS. Experimental results show that VGRIS can effectively schedule GPU resources among various workloads.