Energy-aware Scheduling Research Articles

EAIS: Energy-aware adaptive scheduling for CNN inference on high-performance GPUs

Recently, a large number of convolutional neural network (CNN) inference services have emerged on high-performance Graphic Processing Units (GPUs). However, GPUs are high power consumption units, and the energy consumption increases sharply along with the deployment of deep learning tasks. Although previous studies have considered the latency Service-Level-Objective (SLO) of inference services, they fail to directly take account of the energy consumption. Our investigation shows that coordinating batching and dynamic voltage frequency scaling (DVFS) settings can decrease the energy consumption of CNN inference. But it is affected by (i) larger configuration spaces; (ii) GPUs’ underutilization while data are transferred between CPUs and GPUs; (iii) fluctuating workloads. In this paper, we propose EAIS, an energy-aware adaptive scheduling framework that is comprised of a performance model, an asynchronous execution strategy, and an energy-aware scheduler. The performance model provides valid information about the performance characteristics of CNN inference services to shrink the feasible configuration space. The asynchronous execution strategy overlaps data upload and GPU execution to improve the system processing capacity. The energy-aware scheduler adaptively coordinates batching and DVFS according to fluctuating workloads to minimize energy consumption while meeting latency SLO. Our experimental results on NVIDIA Tesla M40 and V100 GPUs show that, compared to the state-of-the-art methods, EAIS decreases the energy consumption by up to 28.02% and improves the system processing capacity by up to 7.22% while meeting latency SLO. Besides, EAIS has been proved to have good versatility under different latency SLO constraints.

CyEnSe: Cyclic energy-aware scheduling for energy-harvested embedded systems

New advances in energy harvesting from human's body enable battery-less wearable and implantable devices to have more advanced functionalities and hence higher number of tasks to be scheduled on the device. The nature of variable and cyclic energy harvesting from periodic body movements (e.g: walking, heartbeat) challenges the tasks scheduling in these real-time systems, especially for the safety-critical applications when the tasks deadlines are hard. For the first time, this paper addresses cyclic energy harvesting in an offline cyclic-executive scheduling. The paper defines CyEnSe, a scheduling problem in presence of cyclic energy. It proposes two methods, a heuristic, and a linear-programming to solve the hard-deadline periodic task scheduling problem while considers variable cyclic energy arrivals from the energy harvester. We present a comparison among the proposed methods and a First-Fit approach from different optimization point of views, for instance, task utilization, number of tasks, and amount of harvested energy.

Energy-Aware Scheduling for Tasks with Target-Time in Blockchain based Data Centres

Cloud computing infrastructures have intended to provide computing services to end-users through the internet in a pay-per-use model. The extensive deployment of the Cloud and continuous increment in the capacity and utilization of data centers (DC) leads to massive power consumption. This intensifying scale of DCs has made energy consumption a critical concern. This paper emphasizes the task scheduling algorithm by formulating the system model to minimize the makespan and energy consumption incurred in a data center. Also, an energy-aware task scheduling in the Blockchain-based data center was proposed to offer an optimal solution that minimizes makespan and energy consumption. The established model was analyzed with a target-time responsive precedence scheduling algorithm. The observations were analyzed and compared with the traditional scheduling algorithms. The outcomes exhibited that the developed solution incurs better performance with a response to resource utilization and decreasing energy consumption. The investigation revealed that the applied strategy considerably enhanced the effectiveness of the designed schedule.

Clustering-based Energy-aware Scheduling of Smart Residential Area

Clustering-based Energy-aware Scheduling of Smart Residential Area

Reliable Energy-Aware Scheduling Algorithm With Multi-Level Budget for Real-Time Embedded System

Energy consumption of embedded applications has rapidly increased with the advancement of technology and computing. There is a little improvement in energy consumption as compared to computing and storage capacity. Although computing performance has been continuously increasing, power/energy consumption is more critical in the design of real-time embedded systems. Real-time embedded applications need a power management technique to judicially balance the energy consumption and computing performance. It should be done in such a way that the system performance improves along with an increase in the lifespan of the system. The proposed methodology presented in this paper deals with the minimization of energy for time-critical embedded applications. Simulation studies, along with theoretical analysis, have been carried out to show the effectiveness of the proposed three-phase reliable energy-aware scheduling method. It is observed that the proposed approach provides better tolerance (approximately four times) and consumes less energy (35% to 45%) for a wide range of applications.

An energy-aware scheduling of dynamic workflows using big data similarity statistical analysis in cloud computing

An energy-aware scheduling of dynamic workflows using big data similarity statistical analysis in cloud computing

Automated Testbench for Hybrid Machine Learning-Based Worst-Case Energy Consumption Analysis on Batteryless IoT Devices

Batteryless Internet-of-Things (IoT) devices need to schedule tasks on very limited energy budgets from intermittent energy harvesting. Creating an energy-aware scheduler allows the device to schedule tasks in an efficient manner to avoid power loss during execution. To achieve this, we need insight in the Worst-Case Energy Consumption (WCEC) of each schedulable task on the device. Different methodologies exist to determine or approximate the energy consumption. However, these approaches are computationally expensive and infeasible to perform on all type of devices; or are not accurate enough to acquire safe upper bounds. We propose a hybrid methodology that combines machine learning-based prediction on small code sections, called hybrid blocks, with static analysis to combine the predictions to a final upper bound estimation for the WCEC. In this paper, we present our work on an automated testbench for the Code Behaviour Framework (COBRA) that measures and profiles the upper bound energy consumption on the target device. Next, we use the upper bound measurements of the testbench to train eight different regression models that need to predict these upper bounds. The results show promising estimates for three regression models that could potentially be used for the methodology with additional tuning and training.

A Multi-Agent Framework to Solve Energy-Aware Unrelated Parallel Machine Scheduling Problems with Machine-Dependent Energy Consumption and Sequence-Dependent Setup Time

Nowadays,the manufacturing industry consumes almost half of the energy produced on the planet. Within this industry, sustainable scheduling represents a main issue due to the energy efficiency that it is able to achieve. Centralized or decentralized approaches to sustainable scheduling can be found in state-of- the-art techniques using both optimal or heuristic methods. In this paper, a distributed system modelled as a multi-agent framework is proposed to solve energy-aware scheduling problems by applying it to a real case of the injection air moulding industry, which has intense energy consumption. In this proposal, a set of agents collaborate to reach an agreement in order to minimize several objective functions based on: total job weighted tardiness, total setup time, and energy consumption. The proposed system is compared to centralized and distributed approaches, highlighting advantages and drawbacks and providing an interesting point of view about how to tackle energy-aware scheduling with new agreement techniques in a distributed environment.

Energy and SLA-driven MapReduce Job Scheduling Framework for Cloud-based Cyber-Physical Systems

Energy consumption minimization of cloud data centers (DCs) has attracted much attention from the research community in the recent years; particularly due to the increasing dependence of emerging Cyber-Physical Systems on them. An effective way to improve the energy efficiency of DCs is by using efficient job scheduling strategies. However, the most challenging issue in selection of efficient job scheduling strategy is to ensure service-level agreement (SLA) bindings of the scheduled tasks. Hence, an energy-aware and SLA-driven job scheduling framework based on MapReduce is presented in this article. The primary aim of the proposed framework is to explore task-to-slot/container mapping problem as a special case of energy-aware scheduling in deadline-constrained scenario. Thus, this problem can be viewed as a complex multi-objective problem comprised of different constraints. To address this problem efficiently, it is segregated into three major subproblems (SPs), namely, deadline segregation, map and reduce phase energy-aware scheduling. These SPs are individually formulated using Integer Linear Programming. To solve these SPs effectively, heuristics based on Greedy strategy along with classical Hungarian algorithm for serial and serial-parallel systems are used. Moreover, the proposed scheme also explores the potential of splitting Map/Reduce phase(s) into multiple stages to achieve higher energy reductions. This is achieved by leveraging the concepts of classical Greedy approach and priority queues. The proposed scheme has been validated using real-time data traces acquired from OpenCloud. Moreover, the performance of the proposed scheme is compared with the existing schemes using different evaluation metrics, namely, number of stages, total energy consumption, total makespan, and SLA violated. The results obtained prove the efficacy of the proposed scheme in comparison to the other schemes under different workload scenarios.

Energy-Aware Scheduling Based on Marginal Cost and Task Classification in Heterogeneous Data Centers

The energy consumption problem has become a bottleneck hindering further development of data centers. However, the heterogeneity of servers, hybrid cooling modes, and extra energy caused by system state transitions increases the complexity of the energy optimization problem. To deal with such challenges, in this paper, an Energy Aware Task Scheduling strategy (EATS) utilizing marginal cost and task classification method is proposed that cooperatively improves the energy efficiency of servers and cooling systems. An energy consumption model for servers, cooling systems, and state transition is developed, and the energy optimization problem in data centers is formulated. The concept of marginal cost is introduced to guide the task scheduling process. The task classification method is incorporated with the idea of marginal cost to further improve resource utilization and reduce the total energy consumption of data centers. Experiments are conducted using real-world traces, and energy reduction results are compared. Results show that EATS achieves more energy-savings of servers, cooling systems, state transition in comparison to the other two techniques under a various number of servers, cooling modules and task arrival intensities. It is validated that EATS is effective at reducing total energy consumption and improving the resource utilization of data centers.

Empirical power consumption characterization and energy aware scheduling in data centers

Energy-efficient management is key to reduce operational cost and environmental contamination in modern data centers. Energy management and renewable energy utilization are strategies to optimize energy consumption in high-performance computing. In any case, understanding the power consumption behavior of physical servers in datacenter is fundamental to implement energy-aware policies effectively. These policies should deal with possible performance degradation of applications to ensure quality of service. This manuscript presents an empirical evaluation of power consumption for scientific computing applications in multicore systems. Three types of applications are studied, in single and combined executions on Intel and AMD servers, for evaluating the overall power consumption of each application. The main results indicate that power consumption behavior has a strong dependency with the type of application. Additional performance analysis shows that the best load of the server regarding energy efficiency depends on the type of the applications, with efficiency decreasing in heavily loaded situations. These results allow formulating models to characterize applications according to power consumption, efficiency, and resource sharing, which provide useful information for resource management and scheduling policies. Several scheduling strategies are evaluated using the proposed energy model over realistic scientific computing workloads. Results confirm that strategies that maximize host utilization provide the best energy efficiency.

Energy-aware workflow task scheduling in clouds with virtual machine consolidation using discrete water wave optimization

The scientific workflows are high-level complex applications that demand more computing power. The cloud data center (CDC) remains one of the essential models of economic infrastructure for workflow applications. These CDCs consume a lot of electric power while running workflow applications. Hence, efficient energy-aware scheduling techniques are required to perform the task to a virtual machine (VM) mapping. The existing researches overlooked to join the workflow scheduling and VM consolidation which addresses resource utilization and energy consumption effectively. In this article, we propose an energy-aware algorithm for workflow scheduling in cloud computing with VM consolidation called EASVMC. The proposed EASVMC approach is modeled to address the multi-objectives such as energy consumption, resource utilization, and VM migrations. The EASVMC algorithm runs in two phases task scheduling and VM consolidation (VMC). In the first phase, the task with maximum execution length is mapped to the virtual machine that will perform it with the minimum energy. The second phase contains VM consolidation is a prominent NP-hard problem. The VMC phase categorizes the physical hosts into the normal load, under-loaded and overloaded hosts based on CPU utilization. Double threshold values are used for this purpose. VMs from underloaded and overloaded hosts are migrated to normally loaded hosts. For the VMC phase, we used a nature-inspired meta-heuristic approach called the Water Wave Optimization (WWO) algorithm, which finds a suitable migration plan to reduce the energy consumption by increasing the overall resource utilization and switch off idle hosts after migrating its VMs to a suitable target host. The efficiency of our proposed method evaluated using the WorkflowSim simulation tool with five different real-world scientific workloads. The experimental results show that the EASVMC approach surpassed the similar works in stated objectives irrespective of diverse workloads.

Deadline Aware Energy-Efficient Task Scheduling Model for a Virtualized Server

The demand of cloud-based services is increasing day by day. To fulfill the seamless computing demand of consumers, a large number of servers are installed in the datacenters. These servers consume high energy. Reducing the energy consumption of servers with the same performance of service is a vital challenge. Energy-efficient task scheduling is one of the method to reduce energy consumption while preserving the task constraints. This paper presents a deadline and energy-aware scheduling (DEAS) model for a virtualized server to achieve energy efficiency while executing deadline conscious tasks. These tasks are independent in nature and arrives dynamically. The presented DEAS model follows the heuristic approach where the first instance guarantee ratio (GR) is maximized to raise the energy efficiency of the server per unit of work performed. Task slack time is utilized in an innovative way to improve the GR of the server. In this way, the average energy consumption (energy consumption per task) of the server per unit of work done is minimized. The energy efficiency of the server in its idle state is further achieved by applying core-level granularity of dynamic voltage and frequency scaling (DVFS) technology. The DEAS model is evaluated through extensive simulation experiments using the CloudSim simulator. Results show that DEAS model performs better than existing models on the account of considered performance metrics, i.e., GR, total energy consumption, energy consumption per task, and resource utilization.

Dynamic power management under the RUN scheduling algorithm: a slack filling approach

Effective energy-aware scheduling is paramount for current and future multiprocessor real-time systems, being Dynamic Power Management (DPM) one of the employed techniques. In this paper we extend the Reduction to Uniprocessor (RUN) algorithm making it DPM-compliant. RUN is an optimal multiprocessor real-time scheduling for periodic implicit-deadline tasks and it is known to generate low overhead in terms of preemptions and migrations. It is based on an off-line reduction of the target multiprocessor system into one or more uniprocessor systems. On-line scheduling decisions for the latter is then efficiently translated back to the original system. The developed approach in this paper, called Dynamic Slack Filling (DSF-RUN), extends RUN in two aspects. First, the RUN reduction process is adapted to take into account slack entities, properly defined to represent processor spare capacity for DPM management. Second, on-line scheduling rules of RUN are modified so as to generate long time periods in the schedule by dynamically inserting slacks. Our approach has the same off-line complexity of the original RUN algorithm and takes either quadratic or linear on-line complexity in the number of tasks, depending on the strategy used to compute the maximum allowed idle periods. Results obtained with simulations show that our approach consumes less than 10% more energy than an idealized optimum DPM strategy.

Optimizing Energy Consumption of Mobile Games

Games are energy-intensive applications on mobile devices. Optimizing the energy efficiency of games is hence critical for battery-limited mobile devices. Although the advent of energy-aware scheduling (EAS) integrated in recent devices has provided opportunities for improved energy management, the framework is not specifically tuned for game applications. In this paper, we aim to improve the energy efficiency of game applications running on EAS-enabled mobile devices. To this end, we first analyze the functional characteristics of games, and investigate the source of the energy inefficiency. We then propose a scheme, called System-level Energy-optimization for Game Applications (SEGA), to improve the energy efficiency of games. SEGA governs CPU and GPU power consumption in a tightly coupled manner by employing three key techniques: (1) Lsync-aware GPU DVFS governor, (2) adaptive capacity clamping, and (3) on-demand touch boosting. We implemented SEGA on the latest Android-based smartphones. The evaluation results for 23 popular games showed that SEGA reduced the energy consumption of the Google Pixel 2 XL and Samsung Galaxy S9 Plus smartphones, at the device level, by 6.1–22.3 and 4.0–11.7 percent, respectively, with a quality of service (QoS) degradation of 1.1 and 0.5 percent, on average.

Energy-Aware Scheduling of Streaming Applications on Edge-Devices in IoT-Based Healthcare

The reliance on Network-on-Chip (NoC)-based Multiprocessor Systems-on-Chips (MPSoCs) is proliferating in modern embedded systems to satisfy the higher performance requirement of multimedia streaming applications. Task level coarse grained software pipeling also called re-timing when combined with Dynamic Voltage and Frequency Scaling (DVFS) has shown to be an effective approach in significantly reducing energy consumption of the multiprocessor systems at the expense of additional delay. In this article we develop a novel energy-aware scheduler considering tasks with conditional constraints on Voltage Frequency Island (VFI)-based heterogeneous NoC-MPSoCs deploying re-timing integrated with DVFS for real-time streaming applications. We propose a novel task level re-timing approach called R-CTG and integrate it with non linear programming-based scheduling and voltage scaling approach referred to as ALI-EBAD. The R-CTG approach aims to minimize the latency caused by re-timing without compromising on energy-efficiency. Compared to R-DAG, the state-of-the-art approach designed for traditional Directed Acyclic Graph (DAG)-based task graphs, R-CTG significantly reduces the re-timing latency because it only re-times tasks that free up the wasted slack. To validate our claims we performed experiments on using 12 real benchmarks, the results demonstrate that ALI-EBAD out performs CA-TMES-Search and CA-TMES-Quick task schedulers in terms of energy-efficiency.

Structure-Free General Data Aggregation Scheduling for Multihop Battery-Free Wireless Networks

With advances in wireless power transfer techniques, battery-free wireless sensor networks (BF-WSNs) which can support long-term applications, has been attracting increasing interests in recent years. Unfortunately, the problem of minimum latency aggregation scheduling (MLAS) is not well studied in BF-WSNs. Existing works always have a rigid assumption that there is only one single query which is targeted at the whole network. Aiming at making the work more practical and general, we investigate the general MLAS problem in BF-WSNs, which is targeted at any subset of nodes in the network and aimed for an arbitrary number of aggregation queries. First, the general MLAS problem when there is one single query is studied. To control the number of nodes participating in the aggregation process, a node selection algorithm is proposed to cover and connect the whole target nodes. Then, a latency and energy aware scheduling algorithm is proposed to integrate the construction of aggregation tree with the chosen nodes, and the computation of a conflict-free schedule simultaneously, relying on non-predetermined structures. Second, the general MLAS problem when there is a group of aggregation queries is studied. Through designing some special structures to avoid collisions between both current and existing aggregation schedules, an algorithm without any waiting time is proposed. Additionally, the algorithm under physical interference model and dynamic energy arrival model are also presented. The theoretical analysis and simulation results verify that the proposed algorithms have high performance in terms of latency and energy efficiency.

New Heuristics for Scheduling and Distributing Jobs under Hybrid Dew Computing Environments

Mobile grid computing has been a popular topic for researchers due to mobile and IoT devices’ ubiquity and their evergrowing processing potential. While many scheduling algorithms for harnessing these resources exist in the literature for standard grid computing scenarios, surprisingly, there is little insight into this matter in the context of hybrid‐powered computing resources, typically found in Dew and Edge computing environments. This paper proposes new algorithms aware of devices’ power source for scheduling tasks in hybrid environments, i.e., where the battery‐ and non‐battery‐powered devices cooperate. We simulated hybrid Dew/Edge environments by extending DewSim, a simulator that models battery‐driven devices’ battery behavior using battery traces profiled from real mobile devices. We compared the throughput and job completion achieved by algorithms proposed in this paper using as a baseline a previously developed algorithm that considers computing resources but only from battery‐dependent devices called Enhanced Simple Energy‐Aware Schedule (E‐SEAS). The obtained results in the simulation reveal that our proposed algorithms can obtain up to a 90% increment in overall throughput and around 95% of completed jobs in hybrid environments compared to E‐SEAS. Finally, we show that incorporating these characteristics gives more awareness of the type of resources present and can enable the algorithms to manage resources more efficiently in more hybrid environments than other algorithms found in the literature.

Energy-efficient scheduling of streaming applications in VFI-NoC-HMPSoC based edge devices

Energy-aware high-performance computing is becoming a challenging facet for streaming applications at edge devices in Internet-of-Things (IoT) due to the high computational complexity involved. Therefore, the number of processors has increased significantly on the multiprocessor system subsequently, Voltage Frequency Island (VFI) recently adopted for an effective energy management mechanism in the large scale multiprocessor chip designs. In this paper, energy-aware scheduling of real-time streaming applications on edge-devices is investigated. First, an innovative re-timing based technique is developed to transform the dependent workload into an independent task model to avail resources and the wasted slack in the processors with a possible minimal prologue. Moreover, unlike the existing population-based optimization algorithms, a novel population-based algorithm, ARSH-FATI is proposed that can dynamically switch between explorative and exploitative search modes at run-time for performance trade-off. Finally, a communication contention-aware Earliest Edge Consistent Deadline First (EECDF) scheduling algorithm is presented. Our static scheduler ARHS-FATI collectively performs task mapping and ordering. Consequently, its performance is superior to the existing state-of-the-art approach proposed for homogeneous VFI based MPSoCs.

Energy-aware scheduling in edge computing with a clustering method

With the development of Cloud and 5G technology, edge devices have been widely used in various areas. However, the limited battery energy and processing ability of edge devices hinder the usage scope of those devices. Prior studies have typically managed to immigrate virtual machines (VMs) or offload tasks to reduce energy consumption and shorten execution time. In our work, we consider devices that can obtain energy from a green energy source (such as wind energy, or solar energy). First, we use a clustering method to divide nodes into some clusters, each with some edge nodes to ensure the clusters have a minimum distance (defined by energy transferring attenuation ratios between nodes) between nodes in the cluster; the cluster center is a node with a VM. Then, based on the clustering method, a scheduling heuristic is proposed to transfer energy, immigrate VM, and allocate tasks. The simulation result shows that our proposed method reduces both the total energy consumption and the energy consumption from outside of the system (ECFO).