Power-aware Scheduling Research Articles

Discovering the Properties of a Problem of Scheduling Battery Charging Jobs to Minimize the Total Time with the Use of Harmonic Numbers

In this work, we consider a problem from the field of power-aware scheduling in which a fleet of electric vehicles have to be charged in a minimum time. Each vehicle is equipped with a lithium-ion battery of a given capacity. The initial power used for charging each battery is known, whereas it is assumed that the power drops to zero at the moment when the battery gets fully loaded. The power usage function is linear and decreasing. The charging jobs are nonpreemptable and independent, whereas the total available amount of power is limited. The objective is to minimize the schedule length. In this paper, we analyze the case of a problem with identical jobs that already cover a wide variety of practical situations. By employing inverses of natural numbers, similar to harmonic series, we prove two properties of this case, and we also discuss the phenomenon of the stabilization of the difference between the start times of two successive jobs in a schedule. We also take under examination a few special cases of the problem. Some conclusions and directions for future research are given.

Power-aware scheduling of data-flow hardware circuits with symbolic control

Power-aware scheduling of data-flow hardware circuits with symbolic control

Macro Modelling Approach in Power Aware Scheduling

Macro Modelling Approach in Power Aware Scheduling

Cost-sensitive Tensor-based Dual-stage Attention LSTM with Feature Selection for Data Center Server Power Forecasting

Power forecasting has a guiding effect on power-aware scheduling strategies to reduce unnecessary power consumption in data centers. Many metrics related to power consumption can be collected in physical servers, such as the status of CPU, memory, and other components. However, most existing methods empirically exploit a small number of metrics to forecast power consumption. To this end, this article uses feature selection based on causality to explore the metrics that strongly influence the power consumption of different tasks. Moreover, we propose a tensor-based dual-stage attention LSTM to forecast the non-linear and non-periodic power consumption. In the proposed model, a multi-way delay embedding transform is utilized to convert the time series into tensors along the temporal direction. The LSTM combines with the tensor technique and the attention mechanism to capture the temporal pattern effectively. In addition, we adopt the cost-sensitive loss function to optimize the specific power forecasting problem in data centers. The experimental results demonstrate that our method can achieve up to 1.4% to 4.3% forecasting accuracy improvement compared with the state-of-the-art models.

Power and Performance Evaluation of Memory-Intensive Applications

In terms of power and energy consumption, DRAMs play a key role in a modern server system as well as processors. Although power-aware scheduling is based on the proportion of energy between DRAM and other components, when running memory-intensive applications, the energy consumption of the whole server system will be significantly affected by the non-energy proportion of DRAM. Furthermore, modern servers usually use NUMA architecture to replace the original SMP architecture to increase its memory bandwidth. It is of great significance to study the energy efficiency of these two different memory architectures. Therefore, in order to explore the power consumption characteristics of servers under memory-intensive workload, this paper evaluates the power consumption and performance of memory-intensive applications in different generations of real rack servers. Through analysis, we find that: (1) Workload intensity and concurrent execution threads affects server power consumption, but a fully utilized memory system may not necessarily bring good energy efficiency indicators. (2) Even if the memory system is not fully utilized, the memory capacity of each processor core has a significant impact on application performance and server power consumption. (3) When running memory-intensive applications, memory utilization is not always a good indicator of server power consumption. (4) The reasonable use of the NUMA architecture will improve the memory energy efficiency significantly. The experimental results show that reasonable use of NUMA architecture can improve memory efficiency by 16% compared with SMP architecture, while unreasonable use of NUMA architecture reduces memory efficiency by 13%. The findings we present in this paper provide useful insights and guidance for system designers and data center operators to help them in energy-efficiency-aware job scheduling and energy conservation.

A fire detection model based on power-aware scheduling for IoT-sensors in smart cities with partial coverage

Fire detection techniques have received considerable critical attention over the past ten years. Regardless of the progress in the area of fire detection, questions have been raised about the cost, complexity, consumed power from a large number of sensors to analyze sensors’ data. Debate continues about the best strategies for the management of consumed power and how to accelerate the processing of real-time data in fire detection through the internet of things. This paper presents a partial coverage and a power-aware IoT-based fire detection model with different multi-functional sensors in smart cities. In the proposed model, the sleep scheduling approach used for saving sensors energy. This approach will significantly help in saving consumed energy and thus the need for any extra number of nodes required for continuously covering the target area. Moreover, fog computing is applied to process real-time data aggregated from a large number of sensors for running systems more efficiently. Validation of the proposed model was carried out via simulation and experimental testbed implementation with Arduino, sensors, and Raspberry pi. The results obtained indicate how the proposed technique can efficiently determine the sensors to meet the constraints imposed. The most striking finding to emerge from this experimental and simulation study is that the proposed technique helps in ensuring excellent performance in terms of the number of active nodes and the network lifetime, over other state-of-the-art techniques. The most striking finding of this work compared to MWSAC and PCLA, while covering the same area, is the minimization of the number of active nodes by 64.33% and 15%. It raises the network’s lifetime by 91.32% compared to MWSAC and 12% compared to PCLA, respectively.

Dynamic power-aware scheduling of real-time tasks for FPGA-based cyber physical systems against power draining hardware trojan attacks

The present era has witnessed deployment of reconfigurable hardware or field-programmable gate arrays (FPGAs) in diverse domains like automation and avionics, which are cyber physical in nature. Such cyber physical systems are associated with strict power budgets. Efficient real-time task-scheduling strategies exist that ensure execution of maximum number of tasks within the power budget. However, these do not consider hardware threats into account. Recent literature has exposed the existence of hardware trojan horses (HTHs). HTHs are malicious circuitry that remain dormant during testing and evade detection, but get activated at runtime to jeopardize operations. HTHs can be etched into the FPGA fabric by adversaries in the untrustworthy foundries, during fabrication of the FPGAs. Even vendors selling reconfigurable intellectual properties or bitstreams that configure the FPGA fabric for task operation may insert HTHs during writing the bitstream codes. HTHs may cause a variety of attacks which may affect the basic security primitives of the system like its integrity, confidentiality or availability. In this work, we explore how power draining ability of HTHs may reduce lifetime of the system. A self-aware approach is also proposed which detects the affected resources of the system and eradicates their use in future to facilitate system reliability. An offline–online scheduling strategy is proposed for periodic tasks which can ensure reliability of their operations till the expected lifetime of the system. Accommodating non-periodic tasks in the periodic task schedule based on available power is also focused. For experimentation, we consider tasks associated with EPFL benchmarks and demonstrate results based on the metric task success rate for periodic tasks and metric task rejection rate for non-periodic tasks.

Peak-Power-Aware Primary-Backup Technique for Efficient Fault-Tolerance in Multicore Embedded Systems

Multicore platforms offer great potential for task-level redundancy to achieve a degree of fault-tolerance/reliability in embedded systems by exploiting the idle cores. However, due to the Thermal Design Power (TDP) constraint, it may not be possible to simultaneously power-on all cores in a multicore chip at the full-throttle (e.g., in ARM's big.LITTLE architecture). Since TDP is the maximum sustainable power that a chip can dissipate safely (as per the specifications given by a chip vendor), violating TDP triggers a performance throttling mechanism (e.g., by lowering the operating voltage and frequency, or by power-gating with task migration) to avoid possible overheating problems. This can significantly affect the timeliness of the system, and hence, represents a serious challenge in using (off-the-shelf) multicore platforms in real-time embedded systems when exploiting it for full-scale reliability. That means only a few tasks can be afforded to run in a fully reliable mode under a given TDP constraint. In this article, at first, we study the power consumption of task-level redundancy running on multicore platforms. Then, to tackle the peak power problem, we propose a novel primary-backup scheme for power-aware scheduling of real-time tasks on core pairs in multicore systems. The proposed scheme aims at removing overlaps of peak power of concurrently executing tasks to keep the power consumption below the chip-level TDP constraint. This would facilitate higher reliability levels within a given power budget. To do this, considering the tasks' power profiles, we propose a task partitioning method along with maximum-peak-power-first (MPPF) and maximum-peak-power-last (MPPL) policies to schedule original and redundant copies of tasks, respectively. Our experiments show that our technique provides up to 50% (on average by 29.5%) peak power reduction compared to state-of-the-art schemes, while providing the same reliability level.

RETRACTED ARTICLE:Optimal coverage along with connectivity maintenance in heterogeneous wireless sensor network

Heterogeneous wireless sensor network (HWSN) is a network which contains sensor nodes with dissimilar capabilities, such as varying energy, sensing range etc. In Heterogeneous wireless sensor network, Coverage along with connectivity is one of the major issues. Coverage with no connectivity is meaningless in wireless sensor networks. In most of the existing work, coverage and connectivity is not provided efficiently for HWSN. The existing methods used in HWSN to attain coverage and connectivity are static and not applied in the dynamic environment. Optimal Coverage and connectivity maintenance not provided. Energy consumption should be concentrated to maximize the network lifetime. Preserving network connectivity at the same time maximizing coverage by using the less number of energy constrained sensor nodes is the most serious problem. To overcome these issues, in this work we put forward a technique called Coverage Scheduling and Power Aware Connectivity (CSPAC). Grouping the sensors into cluster by the strategy of Dynamic cluster formation. Identifying the minimum active set to provide Q-coverage by Artificial Bee Colony Optimization technique and maximize lifetime by mapping Power Aware Scheduling and Battery Discharge Value. Then Optimal Connectivity Management Technique for reachability to provide strong connectivity. By simulation results, we give you an idea that the proposed technique enhances the network coverage and connectivity in HWSN.

ILP Based Power-Aware Test Time Reduction Using On-Chip Clocking in NoC Based SoC

Network-on-chip (NoC) based system-on-chips (SoC) has been a promising paradigm of core-based systems. It is difficult and challenging to test the individual Intellectual property IP cores of SoC with the constraints of test time and test power. By reusing the on-chip communication network of NoC for the testing of different cores in SoC, the test time and test cost can be reduced effectively. In this paper, we have proposed a power-aware test scheduling by reusing existing on-chip communication network. On-chip test clock frequencies are used for power efficient test scheduling. In this paper, an integer linear programming (ILP) model is proposed. This model assigns different frequencies to the NoC cores in such a way that it reduces the test time without crossing the power budget. Experimental results on the ITC’02 benchmark SoCs show that the proposed ILP method gives up to 50% reduction in test time compared to the existing method.

Power-aware scheduling of real-time applications onto MPSoC platforms with multi-bank shared memory

The on-chip memory architecture of many MPSoC (multiprocessor systems-on-chip) platforms is based on shared, multi-banked scratchpad memories (SPMs). Determining on which SPM bank the data should be placed is critical for the performance of an application. Moreover, data placement and application scheduling/mapping decisions are tightly coupled and should be co-optimized. This work proposes a method for optimal static scheduling, mapping, and data placement on MPSoC platforms with multi-bank memory, considering also energy and time constraints. This problem is addressed with an ILP (Integer Linear Programming) formulation, which is solved using an efficient procedure that omits many constraints from the formulation and adds those omitted constraints only when needed, i.e., when they are violated. Experimental results on classical co-synthesis benchmarks and on a real-world application show that our method is able to find optimal solutions for applications with up to 110 jobs and 27 memory blocks running on a platform with 3 heterogeneous processors and 3 SPM banks of different sizes.

Adaptive Resource Allocation and Provisioning in Multi-Service Cloud Environments

Minimizing the energy consumption and resource usage in cloud computing environment is one of the key research issues. Energy aware resource allocation is used to optimize the power consuming by computer resources and storage in cloud. The proposed system is to improve the utilization of computing resources and reduce energy consumption under workload independent quality of service constraints. Using migration for minimizing the number of active physical nodes the dynamic single threshold VM consolidation leverages fine-grained fluctuations in the workloads and continuously reallocates VMs . A genetic algorithm based power-aware scheduling of resource allocation (G-PARS) has been proposed to solve the dynamic virtual machine allocation policy problem. The experiment results show that strategy that has been proposed has a better performance than other strategies, not only in high Quality Of Service(QoS) but also in less energy consumption.

TA-MCF: Thermal-Aware Fluid Scheduling for Mixed-Criticality System

Fluid scheduling allows tasks to be allocated with fractional processing capacity, which significantly improves the schedulability performance. For dual-criticality systems (DCS), dual-rate fluid-based scheduling has been widely studied, e.g., the state-of-the-art approaches mixed-criticality fluid scheduling (MCF) and MC-Sort. However, most of the existing works on DCS either only focus on the schedulability analysis or minimize the energy consumption treating leakage power as a constant. To this end, this paper considers the effect of temperature on leakage power and proposes a thermal and power aware fluid scheduling strategy, referred to as thermal and energy aware (TA)-MCF which minimizes both the energy consumption and temperature, while ensuring a comparable schedulability ratio compared with the MCF and MC-Sort. Extensive experiments validate the efficiency of TA-MCF.

Hybrid Artificial Bee Colony and Tabu Search Based Power Aware Scheduling for Cloud Computing

Hybrid Artificial Bee Colony and Tabu Search Based Power Aware Scheduling for Cloud Computing

Power Aware Task Scheduling in Compute Cloud

Power Aware Task Scheduling in Compute Cloud

A new rule-based power-aware job scheduler for supercomputers

The fast processing speeds of the current generation of supercomputers provide a great convenience to scientists dealing with extremely large data sets. The next generation of exascale supercomputers could provide accurate simulation results for the automobile industry, aerospace industry, and even nuclear fusion reactors for the very first time. However, the energy cost of super-computing is extremely high, with a total electricity bill of 9 million dollars per year. Thus, conserving energy and increasing the energy efficiency of supercomputers have become critical in recent years. Many researchers have studied this problem and are trying to conserve energy by incorporating the dynamic voltage frequency scaling technique into their methods. However, this approach is limited, especially when the workload is high. In this paper, we developed a power-aware job scheduler by applying a rule-based control method and taking into consideration real-world power and speedup profiles to improve power efficiency while adhering to predetermined power constraints. The intensive simulation results showed that our proposed method is able to achieve the maximum utilization of computing resources as compared to baseline scheduling algorithms while keeping the energy cost under the threshold. Moreover, by introducing a power performance factor based on the real-world power and speedup profiles, we are able to increase the power efficiency by up to 75%.

Prediction and characterization of application power use in a high‐performance computing environment

Power use in data centers and high‐performance computing (HPC) facilities has grown in tandem with increases in the size and number of these facilities. Substantial innovation is needed to enable meaningful reduction in energy footprints in leadership‐class HPC systems. In this paper, we focus on characterizing and investigating application‐level power usage. We demonstrate potential methods for predicting power usage based on a priori and in situ characteristics. Finally, we highlight a potential use case of this method through a simulated power‐aware scheduler using historical jobs from a real scientific HPC system.

Optimizing power consumption in multicore smartphones

This paper addresses the issue of managing power consumption in multicore smartphones via a middleware layer that schedules optimal number of cores for currently running applications taking into account the tradeoff between power consumption, performance and user experience. The paper first describes a simple and accurate method to measure the overall power consumption and then studies the impact of scheduling seven different popular applications over one to four cores on the overall power consumption. Based on this study, the paper proposes three new power-aware scheduling algorithms that dynamically schedule optimal number of cores as well as dynamically adjust the voltage frequency of each online core to achieve the best tradeoff between power consumption, application performance and user experience under the current context. Evaluation from a prototype implementation of the middleware on a quad-core HTC One shows that these algorithms result in significant reduction in power consumption while ensuring good performance and user experience.

Node Scaling Analysis for Power-Aware Real-Time Tasks Scheduling

Multi-core processors achieve a trade-off between the performance and the power consumption by using Dynamic Voltage Scaling (DVS) techniques. In this paper, we study the power efficient scheduling problem of real-time tasks in an identical multi-core system, and present Node Scaling model to achieve power-aware scheduling. We prove that there is a bound speed which results in the minimal power consumption for a given task set, and the maximal value of task utilization, <inline-formula><tex-math notation="LaTeX">$u_{max}$</tex-math></inline-formula> , in a task set is a key element to decide its minimal power consumption. Based on the value <inline-formula><tex-math notation="LaTeX">$u_{max}$ </tex-math></inline-formula> , we classify task sets into two categories: the bounded task sets and the non-bounded task sets, and we prove the lower bound of power consumption for each type of task set. Simulations based on Intel Xeon X5550 and PXA270 processors show Node Scaling model can achieve power efficient scheduling by applying to existing algorithms such as EDF-FF and SPA2. The ratio of power reduction depends on the multi-core processor's property which is defined as the ratio of the bound speed to the maximal speed of the cores. When the ratio of speeds decreases, the ratio of power reduction increases for all the power efficient algorithms.

Power and Fault Aware Reliable Resource Allocation for Cloud Infrastructure

Cloud computing is now trending and more popular in these days for the computation and adopted by many companies like google, amazon, Microsoft etc. As the cloud size increases with increase in number of data center power consumption over a data center increases. As number of request over the data center increase with increase in load and power consumption of the data center. So the requests need to be balanced in such manner which having more effective strategy for resources utilization, request failure and improved power consumption. Cloud computing made it more complicated with respective to requests type that may increase or decrease power consumption. A recent survey on cloud computation shows that the power consumption of a server, increasing in a linear way due to utilization of resource (processors) resulting in request failure at datacenters. Request balancing in such manner without having knowledge of load over server maximize resource utilization but also increasing power consumption at server. So to overcome these issues in cloud Infrastructure as a service (IaaS), we have proposing a fault and power aware scheduling algorithm to minimize the power consumption, request failure and cost over a data center. Proposed algorithm has proven to have better performance in term of load and power efficiency as compared to previously proposed load balancing algorithm for cloud IaaS.