Energy-Efficient Parallel Real-Time Scheduling on Clustered Multi-Core

Bhuiyan, Ashikahmed; Liu, Di; Khan, Aamir; Saifullah, Abusayeed; Guan, Nan; Guo, Zhishan

doi:10.1109/tpds.2020.2985701

Cited by 28 publications

(7 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The existing models need to fit travel characteristics from massive data and simulate the travel and charge-discharge behavior of large-scale EV clusters, which may cause huge calculation burdens. In recent years, the development direction of utilizing CPU computing power has shifted from increasing single-core frequency to balancing multi-core performance [40]- [42]. At the same time, cloud service providers such as Amazon Web Services, Microsoft Azure and Huawei Cloud can provide Elastic Cloud Server (ECS) to meet the need of differentiated computing [43]- [45].…”

Section: G Parallel Implementation Of the Methodsmentioning

confidence: 99%

A Calculation Model of Charge and Discharge Capacity of Electric Vehicle Cluster Based on Trip Chain

Liang

Lee

2020

IEEE Access

View full text Add to dashboard Cite

The rapid response characteristics and high-speed growth of electric vehicles (EVs) demonstrate its potential to provide auxiliary frequency regulation services for independent system operators through vehicle-to-grid (V2G). However, due to the spatiotemporal random dynamics of travel behavior, it is challenging to evaluate the ability of EV cluster to provide ancillary services under the premise of reaching the expected state of charge (SOC) level. To address this issue, a novel calculation model of charge and discharge capacity of EV cluster based on trip chain with excellent parallel computing performance is presented in this work. Following the introduction of the characteristic variables of the proposed trip chain model, the user's continuous travel behavior in a time scale of several weeks is simulated. In particular, a bidirectional V2G scheduling strategy based on the five-zone map is designed to guide the charging and discharging behavior of EVs, where the expected SOC levels are guaranteed. The results of a 3-week travel simulation verify the effectiveness of the presented model in coordinating the V2G scheme and calculating the charge and discharge capacity of the EV cluster.

show abstract

Section: G Parallel Implementation Of the Methodsmentioning

confidence: 99%

A Calculation Model of Charge and Discharge Capacity of Electric Vehicle Cluster Based on Trip Chain

Liang

Lee

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…These VMs are charged based on time unit (hourly pricing model). Using the fact that the recent processors support DVFS techniques, that enable every processor to scale the voltage and frequency at runtime to reduce power utilization, 42 we assume that each VM in the CDC supports DVFS. For example, six different VM instance types available in Amazon EC2: 6 a1.medium, a1.large, a1.xlarge, a1.2xlarge, a1.4xlarge and a1.metal.…”

Section: System Modelsmentioning

confidence: 99%

Energy and cost aware workflow scheduling in clouds with deadline constraint

Medara

Singh

Sompalli

2022

Concurrency and Computation

View full text Add to dashboard Cite

Cloud computing is a promising platform for executing scientific workflow applications.Commonly, the scientific workflow applications are complex in size and computation intensive. Task scheduling in clouds is a familiar NP-complete problem. Efficient workflow application scheduling is critical for meeting multi-objectives in cloud environments. By reason of its pivotal role, this problem has been widely studied, and many approaches have been developed. Most of the algorithms focus on schedule makespan and execution cost. The cloud data centers consume a large amount of energy while running workflow applications due to a lack of efficient scheduling algorithms to perform the task to a virtual machine (VM) mapping. Hence, there is a significant need to address energy utilization in the cloud data center (CDC) for two reasons such as data center operational cost optimization and to improve the environment. This article presents an energy and cost-aware scheduling (ECWS) approach with the objectives of decreasing energy utilization, execution cost, and maximizing utilization of the resources. The ECWS algorithm is a heterogeneous earliest finish time (HEFT) based energy-efficient heuristic for cloud scheduler. The ECWS algorithm includes three sub-algorithms like RE calculation, RE Threshold selection, and slack algorithm.The effectiveness of the proposed algorithm is evaluated on the WorkflowSim tool using distinct workloads from various scientific areas. The experimental outcome exhibit that the proposed algorithm achieved significant energy conservation, maximized resource utilization, and cost-saving when compared with related well-known algorithms.

show abstract

“…Both the work by Bhuiyan et al [22] and Guo et al [3] considered a simplified model (e.g., the number of cores are unlimited, the entire schedule until the hyper-period available a priori) to propose the energy-aware real-time scheduling of DAG tasks. Based on the DAG task model, some recent works have studied the energy-aware scheduling in a homogeneous and heterogeneous clustered platforms [23]- [25]. The work by Saifullah et al [26] studied the CPU energy optimization of the DAG task considering the federated scheduling policy.…”

Section: Related Workmentioning

confidence: 99%

A Multi-Level DPM Approach for Real-Time DAG Tasks in Heterogeneous Processors

Reghenzani

Bhuiyan

Fornaciari

et al. 2021

2021 IEEE Real-Time Systems Symposium (RTSS)

Self Cite

View full text Add to dashboard Cite

The modeling and analysis of real-time applications focus on the worst-case scenario because of their strict timing requirements. However, many real-time embedded systems include critical applications requiring not only timing constraints but also other system limitations, such as energy consumption.In this paper, we study the energy-aware real-time scheduling of Directed Acyclic Graph (DAG) tasks. We integrate the Dynamic Power Management (DPM) policy to reduce the Worst-Case Energy Consumption (WCEC), which is an essential requirement for energy-constrained systems. Besides, we extend our analysis with tasks' probabilistic information to improve the Average-Case Energy Consumption (ACEC), which is, instead, a common non-functional requirement of embedded systems. To verify the benefits of our approach in terms of reduced energy consumption, we finally conduct an extensive simulation, followed by an experimental study on an Odroid-H2 board. Compared to the state-of-the-art solution, our approach is able to reduce the power consumption up to 32.1%.

show abstract

Energy-Efficient Parallel Real-Time Scheduling on Clustered Multi-Core

Cited by 28 publications

References 44 publications

A Calculation Model of Charge and Discharge Capacity of Electric Vehicle Cluster Based on Trip Chain

A Calculation Model of Charge and Discharge Capacity of Electric Vehicle Cluster Based on Trip Chain

Energy and cost aware workflow scheduling in clouds with deadline constraint

A Multi-Level DPM Approach for Real-Time DAG Tasks in Heterogeneous Processors

Contact Info

Product

Resources

About