Dynamic Voltage And Frequency Scaling Research Articles

An Energy-Efficient Graphics Processor in 14-nm Tri-Gate CMOS Featuring Integrated Voltage Regulators for Fine-Grain DVFS, Retentive Sleep, and <inline-formula> <tex-math notation="LaTeX">${V}_{\text{MIN}}$ </tex-math> </inline-formula> Optimization

Graphics workloads make highly dynamic use of resources such as execution units (EUs), and thus can benefit from fast, fine-grain dynamic voltage and frequency scaling (DVFS) and retentive sleep. This paper presents a 14-nm graphics processing unit (GPU) prototype with modified EUs which include an integrated voltage regulator (IVR). The IVR enables energy-efficient EU turbo operation, data retention, and $V_{\text {MIN}}$ optimization per EU. Silicon measurements show that IVR-enabled EU turbo operation offers up to 32% (average 29%) energy reduction at constant performance.

Energy-Efficient Runtime Management of Heterogeneous Multicores using Online Projection

Heterogeneous multicores offer flexibility in the form of different core types and Dynamic Voltage and Frequency Scaling (DVFS), defining a vast configuration space. The optimal configuration choice is not always straightforward, even for single applications, and becomes a very difficult problem for dynamically changing scenarios of concurrent applications with unpredictable spawn and termination times and individual performance requirements. This article proposes an integrated approach for runtime decision making for energy efficiency on such systems. The approach consists of a model that predicts performance and power for any possible decision and low-complexity heuristics that use this model to evaluate a subset of possible decisions to choose the best. The model predicts performance by projecting standalone application profiling data to the current status of the system and power by using a set of platform-specific parameters that are determined only once for a given system and are independent of the application mix. Our approach is evaluated with a plethora of dynamic, multi-application scenarios. When considering best effort performance to be adequate, our runtime achieves on average 3% higher energy efficiency compared to the powersave governor and 2× better compared to the other linux governors. Moreover, when also considering individual applications’ performance requirements, our runtime is able to satisfy them, giving away 18% of the system’s energy efficiency compared to the powersave, which, however, misses the performance targets by 23%; at the same time, our runtime maintains an efficiency advantage of about 55% compared to the other governors, which also satisfy the performance constraints.

Energy-aware dynamic resource management in elastic cloud datacenters

In clouds, placement of on-demand applications on heterogeneous machines, has turned out to be a crucial research problem, particularly, in terms of performance and energy consumption. Techniques like Dynamic Voltage and Frequency Scaling (DVFS), processor speed adjustment and features such as turning off displays, activating sleep modes, etc. are only useful for decreasing the energy consumption of a single machine, at marginal loss in performance. They cannot be used to achieve significant power optimization in High Performance Computing (HPC) systems such as grids, and cloud datacenters; because power saved by scaling down the processor voltage is far less than switching off a machine. Resource management, using dynamic consolidation of VMs, allows cloud service providers to optimize resource usability, performance and decrease power consumption. This paper investigates various resource management techniques, and suggests several heuristic approaches to optimise energy consumption and performance in elastic datacenters. Using real workload datasets, our evaluation suggests that a combination of the proposed VM allocation and consolidation with migration control technique could save approximately 1.96%–9.38% energy, and improve 0.32%–5.96% performance, as compared to its closest rivals.

Evaluation of the impacts of failures and resource heterogeneity on the power consumption and performance of IaaS clouds

In this paper, we model the infrastructure of cloud computing systems to evaluate the power consumption and performance measures using stochastic activity networks (SANs). In the proposed model, servers run different numbers of virtual machines (VMs) and tasks are divided into two categories, namely low and high, demonstrating their priorities to run. Furthermore, both servers and VMs can fail during their operation. We consider the shutting down and the dynamic voltage and frequency scaling (DVFS) techniques for decreasing the power consumption in the proposed model. These techniques can also affect the performance of cloud systems. By evaluating the results obtained from the proposed SAN model in different scenarios, we conclude that the scenario in which servers run different numbers of VMs compared to scenarios at which servers run the same number of VMs is optimal in terms of both power and performance measures. Furthermore, the obtained results represent that failures do not much affect the power consumption, but the failure of servers, in comparison with the failure of VMs, has a great impact on the performance of the system under study. We also cross-validate the results obtained from the proposed analytical model, by applying the Mobius modeling tool, with the simulation results gained from the CloudSim framework.

A smart coordinated temperature feedback controller for energy-efficient data centers

Energy cost plays a big role in the running cost of data centers and cutting it would get great economic and green benefits. This paper presents a coordinated MPC-based (Model Predictive Control) scheduling strategy called Co-ThermoRing to smartly reduce energy cost in data centers. Co-ThermoRing makes use of an online hierarchical feedback control mechanism which is composed by a global cluster-level temperature controller and many local node-level efficiency controllers. The inlet temperatures of all nodes can be kept balanced to minimize the heat recirculation by the temperature controller, and DVFS (Dynamic Voltage and Frequency Scaling) technology is adopted by each efficiency controller to utilize every computing node intelligently. This method can also contain temperature and utilization fluctuations with little performance degradation. With a TPC-W (Transaction Processing Performance Benchmark™ W) run in a real data center to simulate an online bookstore behavior our method is proved to have less energy consumption in normal cases and emergency cases compared with existing solutions such as LRHex, C-Oracle-F and ThermoRing.

Frequency Scaling As a Security Threat on Multicore Systems

Most modern processors use dynamic voltage and frequency scaling (DVFS) for power management. DVFS allows to optimize power consumption by scaling voltage and frequency depending on performance demand. Previous research has indicated that this frequency scaling might pose a security threat in the form of a covert channel, which could leak sensitive information. However, an analysis able to determine whether DVFS is a serious security issue is still missing. In this paper, we conduct a detailed analysis of the threat potential of a DVFS-based covert channel. We investigate two multicore platforms representative of modern laptops and hand-held devices. Furthermore, we develop a channel model to determine an upper bound to the channel capacity, which is in the order of 1 bit per channel use. Last, we perform an experimental analysis using a novel transceiver implementation. The neural network-based receiver yields packet error rates between 1% and 8% at average throughputs of up to 1.83 and 1.20 bps for platforms representative of laptops and hand-held devices, respectively. Considering the well-known small message criterion, our results show that a relevant covert channel can be established by exploiting the behavior of computing systems with DVFS.

VMSAGE: A virtual machine scheduling algorithm based on the gravitational effect for green Cloud computing

The area of sustainable green smart computing highlights key challenges towards reducing cost and carbon dioxide emissions due to the high-energy consumption of Cloud data centres. Here, we focus on the Cloud virtual machine (VM) scheduling that is usually based on simple algorithms, e.g. VM placement on nodes with low memory usage. This approach fails to consider the actual configuration of nodes inside the server rack resulting in local overheating of Cloud data centres. To solve this, we propose a VM scheduling algorithm based on the gravitational effect, called VMSAGE, to optimize energy efficiency of Cloud computing systems. Inspired by the physical gravitation model, we define the thermal repulsion and logical gravitation factors between physical nodes and VMs. To achieve optimized VM scheduling, we propose a gravitation function that refers to the calculation of the logical quality of each VM, host and rack through the algorithm, so as to draw the attractiveness between them. Based on the concept of dimension reduction, VMSAGE conducts the two-dimensional plane target selection twice to reduce the computational cost. Additionally, VMSAGE evaluates attributes of the computer room to carry out the VM deployment. To demonstrate the effectiveness of our solution, we compare it with the Best Fit Heuristic (BFH) and the dynamic voltage and frequency scaling (DVFS) algorithms. The results indicate that our algorithm achieves 10% and 20% optimized energy consumption respectively. The experimental results highlight our contribution, in where VMSAGE can significantly reduce energy consumption rates and VM migration times.

Dynamic Lifetime Reliability Management for Chip Multiprocessors

We introduce an algorithm for dynamic lifetime reliability optimization of chip multiprocessors (CMPs). The proposed dynamic reliability management (DRM) algorithm combines thread migration and dynamic voltage and frequency scaling (DVFS) as the two primary techniques to change the CMP operation. The goal is to increase the lifetime reliability of the overall system to the desired target with minimal performance degradation. We test the proposed algorithm with a variety of benchmarks on 16 and 64 core network-on-chip (NoC) based CMP architectures. Full-system based simulations using a customized GEM5 simulator demonstrate that lifetime reliability can be improved by 100 percent for an average performance penalty of 7.7 and 8.7 percent for the two CMP architectures.

Hot-N-Cold model for energy aware cloud databases

A lot of cloud computing and cloud database techniques are adopted in industry and academia to face the explosion of the arrival of the big data era. Meanwhile, energy efficiency and energy saving become a major concern in data centers, which are in charge of large distributed systems and cloud databases. However, the phenomenon of energy wasting is related to resource provisioning. Hot-N-Cold model is introduced in this paper, which uses workload predictions and DVFS(Dynamic Voltage and Frequency Scaling) to cope with the resource provisioning problem within energy aware cloud database systems. In this model, the resource provisioning problem is considered as two bounded problems. A nonlinear programming algorithm and a multi-phase algorithm are proposed to solve them. The experimental results show that one of the proposed algorithms has great scalability which can be applied to a cloud database system deployed on 70 nodes. Using Hot-N-Cold model can save up to 21.5% of the energy of the running time.

Power mitigation of a heterogeneous multicore architecture on FPGA/ASIC by DFS/DVFS techniques

This article presents an integrated self-aware computing model in a Heterogeneous Multicore Architecture (HMA) to mitigate the power dissipation of an Orthogonal Frequency-Division Multiplexing (OFDM) receiver. The proposed platform consists of template-based Coarse-Grained Reconfigurable Array (CGRA) devices connected through a Network-on-Chip (NoC) around a few Reduced Instruction-Set Computing (RISC) cores. The self-aware computing model exploits Feedback Control System (FCS) which constantly monitors the execution-time of each core and dynamically scales the operating frequency of each node of the NoC depending on the worst execution-time. Therefore, the performance of the overall system is equalized towards a desired level besides mitigating the power dissipation. Measurement results obtained from Field-Programmable Gate Array (FPGA) synthesis show up to 20.2% dynamic power dissipation and 16.8% total power dissipation savings. Since FCS technique can be employed for scaling the frequency and the voltage and on the other hand, voltage supply cannot be scaled on the FPGA-based prototyped platform, the implementation is also estimated in 28nm Ultra-Thin Body and Buried oxide (UTBB) Fully-Depleted Silicon-On-Insulator (FD-SOI) Application-Specific Integrated Circuit (ASIC) technology to scale voltage in addition to frequency and get more benefits in terms of dynamic power dissipation reduction. Subsequent to synthesizing the whole platform on ASIC and scaling the voltage and frequency simultaneously as a Dynamic Voltage and Frequency Scaling (DVFS) method, significant dynamic power dissipation savings by 5.97X against Dynamic Frequency Scaling (DFS) method were obtained.

Learning-Directed Dynamic Voltage and Frequency Scaling Scheme with Adjustable Performance for Single-Core and Multi-Core Embedded and Mobile Systems

Dynamic voltage and frequency scaling (DVFS) is a well-known method for saving energy consumption. Several DVFS studies have applied learning-based methods to implement the DVFS prediction model instead of complicated mathematical models. This paper proposes a lightweight learning-directed DVFS method that involves using counter propagation networks to sense and classify the task behavior and predict the best voltage/frequency setting for the system. An intelligent adjustment mechanism for performance is also provided to users under various performance requirements. The comparative experimental results of the proposed algorithms and other competitive techniques are evaluated on the NVIDIA JETSON Tegra K1 multicore platform and Intel PXA270 embedded platforms. The results demonstrate that the learning-directed DVFS method can accurately predict the suitable central processing unit (CPU) frequency, given the runtime statistical information of a running program, and achieve an energy savings rate up to 42%. Through this method, users can easily achieve effective energy consumption and performance by specifying the factors of performance loss.

DVFS Algorithms of GPU and Memory for Mobile GPGPU Applications: A case study

Although both OpenCL and RenderScript have allowed the General-Purpose Graphics Processing Unit (GPGPU) to be used even for mobile GPUs, it is still difficult for mobile applications to use the GPGPU for several reasons. One of the reasons is that mobile devices place restrictions on GPU performance through power-saving technologies such as Dynamic Voltage and Frequency Scaling (DVFS). DVFS tries to control the balance between performance and energy consumption based on the application’s requirements. This technology has been successful in many cases and is widely used; however, it significantly decreases the performance of GPGPU applications. In this paper, we propose novel DVFS algorithms for GPU and memory when the GPGPU applications run. The suggested algorithms decreased the energy consumption by more than 0.7 times without any algorithm changes, and improved the energy efficiency (performance per watt) by more than 3.42 times in comparison with the conventional interval-based algorithm.

PL-DVFS: combining Power-aware List-based scheduling algorithm with DVFS technique for real-time tasks in Cloud Computing

In recent years, energy efficiency has emerged as one of the most important design requirements for modern computing systems, ranging from single servers to data centers and Clouds, as they continue to consume an enormous amount of electrical power. Cloud computing can be used to achieve energy efficiency through efficient task scheduling in the distributed environment. This efficient task scheduling helps to improve resource utilization, which, in turn, helps to minimize energy consumption. In this paper, we work toward minimizing energy of directed acyclic graph-structured applications on heterogeneous cloud system. The paper also combines power-aware list-based scheduling algorithm with dynamic voltage and frequency scaling (DVFS) technique for real-time tasks (PL-DVFS) to maintain the quality of service while considering tasks deadlines. The goal of the approach is to improve performance and overall reduced energy consumption comprising CPU energy (busy and idle) and communication energy. Experiments conducted with synthetic workflow graphs clearly demonstrate the advantage of the proposed approach.

Energy-Efficient Fault-Tolerant Scheduling of Reliable Parallel Applications on Heterogeneous Distributed Embedded Systems

Dynamic voltage and frequency scaling (DVFS) is a well-known energy consumption optimization technique in embedded systems and dynamically scaling down the voltage of a chip has been developed to achieve energy-efficient optimization. However, this operation may lead to a sharp rise in transient failures of processors and consequently weaken the reliability of systems. Reliability goal is an important functional safety requirement and must be satisfied for safety-critical applications. In this study, we aim to implement energy-efficient fault-tolerant scheduling for a reliable parallel application on heterogeneous distributed embedded systems, where the parallel application is described by a directed acyclic graph (DAG). An energy-efficient scheduling with a reliability goal (ESRG) algorithm is presented to reduce the energy consumption while satisfying the reliability goal for the parallel application. Considering that the application's reliability goal is unreachable if its reliability goal exceeds a certain threshold via ESRG, we further propose an energy-efficient fault-tolerant scheduling with a reliability goal (EFSRG) algorithm to reduce the energy consumption while satisfying the reliability goal based on an active replication scheme. Experimental results confirm that the energy consumption reduced by the proposed EFSRG algorithm is higher than those reduced by other approaches under different scale conditions.

Software and DVFS Tuning for Performance and Energy-Efficiency on Intel KNL Processors

Energy consumption of processors and memories is quickly becoming a limiting factor in the deployment of large computing systems. For this reason, it is important to understand the energy performance of these processors and to study strategies allowing their use in the most efficient way. In this work, we focus on the computing and energy performance of the Knights Landing Xeon Phi, the latest Intel many-core architecture processor for HPC applications. We consider the 64-core Xeon Phi 7230 and profile its performance and energy efficiency using both its on-chip MCDRAM and the off-chip DDR4 memory as the main storage for application data. As a benchmark application, we use a lattice Boltzmann code heavily optimized for this architecture and implemented using several different arrangements of the application data in memory (data-layouts, in short). We also assess the dependence of energy consumption on data-layouts, memory configurations (DDR4 or MCDRAM) and the number of threads per core. We finally consider possible trade-offs between computing performance and energy efficiency, tuning the clock frequency of the processor using the Dynamic Voltage and Frequency Scaling (DVFS) technique.

Thermal-Aware and DVFS-Enabled Big Data Task Scheduling for Data Centers

Big data has received considerable attentions in recent years because of massive data volumes in multifarious fields. Considering various “V” features, big data tasks are usually highly complex and computational intensive. These tasks are generally performed in parallel in data centers resulting in massive energy consumption and Green House Gases emissions. Therefore, efficient resource allocation considering the synergy of the performance and energy efficiency is one of the crucial challenges today. In this paper, we aim to achieve maximum energy efficiency by combining thermal-aware and dynamic voltage and frequency scaling (DVFS) techniques. This paper proposes: (a) a thermal-aware and power-aware hybrid energy consumption model synchronously considering the computing, cooling, and migration energy consumption; (b) a tensor-based task allocation and frequency assignment model for representing the relationship among different tasks, nodes, time slots, and frequencies; and (c) a big data Task Scheduling algorithm based on Thermal-aware and DVFS-enabled techniques (TSTD) to minimize the total energy consumption of data centers. The experimental results demonstrate that the proposed TSTD algorithm significantly outperforms the state-of-the-art energy efficient algorithms from total, computing, and cooling energy consumption perspectives, as well as cooling energy consumption proportion and total energy consumption savings.

Energy conscious multi-site computation offloading for mobile cloud computing

In mobile devices, one of the major reasons for battery consumption is due to computation of complex applications. To provide high performance of execution on the mobile devices, the concept of mobile cloud computing (MCC) is used. MCC allows the computation complex modules to offload onto a cloud from a mobile device, which helps to remove the resource constraint condition of the mobile devices. Computations can also be offloaded to nearby clouds called multi-sites, which may have different resources, access delays, computation capability or service charges. The mobile users can specify their priority such as total completion time, cost or energy saving of the application execution in MCC environment. But most of the existing research is focused to optimize only one objective, i.e., either total completion time or cost or energy. But when offloading computation modules onto cloud-based multi-sites, a tradeoff solution is required to strike a balance between the total completion time and energy savings. Further, the entire computational execution on the cloud is to be served efficiently with optimal power utilization. Various algorithms are developed to reduce power consumption, and one such algorithm is dynamic voltage and frequency scaling (DVFS) algorithm. In this paper, new algorithms known as cost and time constraint task partitioning and offloading algorithm (CTTPO), multi-site task scheduling algorithm (MTS) based on teaching, learning-based optimization and the energy saving on multi-sites (ESM) using DVS technique are proposed. CTTPO deals with trade-off between time and cost for the task partitioning and offloading. The MTS algorithm deals with time efficient scheduling on multi-sites, and ESM algorithm saves the energy on the multi-sites by switching the sites from high voltage to low voltage during ideal time. The simulation study demonstrates that the proposed algorithms outperformed the existing techniques based on time, cost and energy parameters.

Effect of frequency scaling granularity on energy-saving strategies

Energy efficiency and energy proportional computing have become major constraints in the design of modern exascale platforms. Dynamic voltage and frequency scaling (DVFS) is one of the most commonly used and effective techniques to dynamically reduce power consumption based on workload characteristics. Many energy-saving strategies, however, employ DVFS without considering its applicability level in a given multicore processor. The present work demonstrates that disregarding the DVFS granularity as a design constraint while developing energy-saving strategies may reduce the efficacy of the strategy for certain application classes. Specifically, DVFS applicability levels (called granularities) are determined here for three different Intel processor microarchitectures. Then, the degradation in energy savings is evaluated if the DVFS granularity is disregarded. Experiments were conducted on the widely used quantum chemistry packages, General Atomic and Molecular Electronic Structure System (GAMESS) and NWChem, and show that, for granularity unaware energy-saving techniques, the overall energy consumption may increase by as much as 19%.

Energy-Aware Computing for Infrastructure Clouds Using DVFS

With the increasing demand of cloud enabled services, service providers are deploying more and more data centers, resulting in huge amount of energy consumption. According to various surveys, energy is one of the dominant factors that impact the cost of cloud usage. Excessive power consumption leads to large power bills in addition to emission of greenhouse gases. Sufficient amount of energy can be conserved by running services at low power mode and powering off servers that are idle. Different authors have come up with different energy-aware approaches with the aim to minimize overall energy consumption. In this paper we are trying to investigate the difference in energy consumption level when data centre is equipped with energy aware approach i.e., using Dynamic Voltage and Frequency Scaling (DVFS) approach and without using energy efficient approach. Simulation results show that DVFS supported data centers consume 66% less energy as compared to non power aware approaches. Average resource utilization is optimal with DVFS as compared to non power aware approach.

Feedback-Based Admission Control for Firm Real-Time Task Allocation with Dynamic Voltage and Frequency Scaling

Feedback-based mechanisms can be employed to monitor the performance of Multiprocessor Systems-on-Chips (MPSoCs) and steer the task execution even if the exact knowledge of the workload is unknown a priori. In particular, traditional proportional-integral controllers can be used with firm real-time tasks to either admit them to the processing cores or reject in order not to violate the timeliness of the already admitted tasks. During periods with a lower computational power demand, dynamic voltage and frequency scaling (DVFS) can be used to reduce the dissipation of energy in the cores while still not violating the tasks’ time constraints. Depending on the workload pattern and weight, platform size and the granularity of DVFS, energy savings can reach even 60% at the cost of a slight performance degradation.