Energy-performance Trade-off Research Articles

Energy performance tradeoff assessment for varying design parameters in office buildings

ABSTRACT Energy performance and heat release from buildings can differ with variations in envelope and operational parameters. This study presents an evidence-based assessment for different study locations in Japan including Sapporo, Tokyo, and Osaka. The analysis used a comparative approach using a reference energy performance simulation model for large office buildings using EnergyPlus to assess the impact of design variations on building performance. The analysis included 66 simulations with 10 design parameters and 2 construction types. The study coupled energy performance with heat release from the building towards outdoor environments since heat release can influence the local thermal environments. Desirable tradeoffs were observed for timber constructions by increasing the cooling setpoint by 1°C reduced cooling and heating energy use by 10.7% and 11.9% in Sapporo, 8.5% and 22.7% in Tokyo, and 6.8% and 24.4% in Osaka. Similarly, cooling and heating peak loads were reduced by 5.9% and 2.4% in Sapporo, 5.4% and 2.0% in Tokyo, and 6.5% and 5.2% in Osaka. Additionally, nighttime heat release also decreased by 12.4% and 13.6% in Sapporo and Tokyo, and 6.2% and 5.7% in Osaka for concrete and timber constructions. The study provides early-stage design support for energy-efficient buildings and improved local thermal environments.

Simplified Compressor and Encoder Designs for Low-Cost Approximate Radix-4 Booth Multiplier

In this brief, we present a novel design methodology of cost-effective approximate radix-4 Booth multipliers, which can significantly reduce the power consumption of error-resilient signal processing tasks. In contrast that the prior studies only focus on the approximation of either the partial product generation with encoders or the partial product reductions with compressors, the proposed method considers two major processing steps jointly by forcing the generated error directions to be opposite to each other. As the internal errors are naturally balanced to have zero mean, as a result, the proposed approximate Booth multiplier can minimize the required processing energy under the same number of approximate bits compared to the previous designs. Simulation results on FIR filtering and image classification applications reveal that the proposed approximate Booth multiplier shows the most attractive energy-performance trade-offs, achieving 28% and 34% of energy reduction compared to the exact Booth multiplier, respectively, with negligible accuracy loss.

Performance–energy trade-offs of deep learning convolution algorithms on ARM processors

In this work, we assess the performance and energy efficiency of high-performance codes for the convolution operator, based on the direct, explicit/implicit lowering and Winograd algorithms used for deep learning (DL) inference on a series of ARM-based processor architectures. Specifically, we evaluate the NVIDIA Denver2 and Carmel processors, as well as the ARM Cortex-A57 and Cortex-A78AE CPUs as part of a recent set of NVIDIA Jetson platforms. The performance–energy evaluation is carried out using the ResNet-50 v1.5 convolutional neural network (CNN) on varying configurations of convolution algorithms, number of threads/cores, and operating frequencies on the tested processor cores. The results demonstrate that the best throughput is obtained on all platforms with the Winograd convolution operator running on all the cores at their highest frequency. However, if the goal is to reduce the energy footprint, there is no rule of thumb for the optimal configuration.

CHAMP: Channel Merging Process for Cost-Efficient Highly-Pruned CNN Acceleration

This paper presents an advanced offline scheduling scheme to improve the accelerator efficiency, especially for the highly-pruned convolutional neural networks (HP-CNNs). Based on the existing outlier-aware accelerator design, we demonstrate the efficiency drop of HP-CNN processing for the first time, and element-wise channel merging is proposed to make a dense processing sequence even for the highly-pruned model. The dedicated hardware architecture is also presented to process the merged channels with the minimum hardware-level overheads, improving the energy efficiency for handling HP-CNNs by preserving the hardware utilization. We further investigate the optimal size of accumulator and multiplexer, in addition to the number of merged channels, exploiting the attractive energy-performance trade-offs. As a result, unlike the practical HP-CNNs for the on-device solutions, the proposed method enhances the overall efficiency by up to 33% compared to the state-of-the-art schemes.

Look-ahead energy efficient VM allocation approach for data centers

Energy efficiency is an important issue for reducing environmental dissipation. Energy efficient resource provisioning in cloud environments is a challenging problem because of its dynamic nature and varied application workload characteristics. In the literature, live migration of virtual machines (VMs) among servers is commonly proposed to reduce energy consumption and to optimize resource usage, although it comes with essential drawbacks, such as migration cost and performance degradation. Energy efficient provisioning is addressed at the data center level in this research. A novel efficient resource management algorithm for virtualized data centers that optimizes the number of servers to meet the requirements of dynamic workloads without migration is proposed in this paper. The proposed approach, named Look-ahead Energy Efficient VM Allocation (LAA), contains a Holt Winters-based prediction module. Energy efficiency and performance are inversely proportional. The energy-performance trade-off relies on periodic comparisons of the predicted and active numbers of servers. To evaluate the proposed algorithm, experiments are conducted with real-world workload traces from Google Cluster. LAA is compared with the best approach provided by CloudSim based on VM migration called Local Regression-Minimum Migration Time (LR-MMT). The experimental results show that the proposed algorithm leads to a consumption reduction of up to 45% to complete one workload compared with the LR-MMT.

Learning, Computing, and Trustworthiness in Intelligent IoT Environments: Performance-Energy Tradeoffs

An Intelligent IoT Environment (iIoTe) is comprised of heterogeneous devices that can collaboratively execute semi-autonomous IoT applications, examples of which include highly automated manufacturing cells or autonomously interacting harvesting machines. Energy efficiency is key in such edge environments, since they are often based on an infrastructure that consists of wireless and battery-run devices, e.g., e-tractors, drones, Automated Guided Vehicle (AGV)s and robots. The total energy consumption draws contributions from multipleiIoTe technologies that enable edge computing and communication, distributed learning, as well as distributed ledgers and smart contracts. This paper provides a state-of-the-art overview of these technologies and illustrates their functionality and performance, with special attention to the tradeoff among resources, latency, privacy and energy consumption. Finally, the paper provides a vision for integrating these enabling technologies in energy-efficient iIoTe and a roadmap to address the open research challenges

Dynamic performance-Energy tradeoff consolidation with contention-aware resource provisioning in containerized clouds.

Containers have emerged as a more portable and efficient solution than virtual machines for cloud infrastructure providing both a flexible way to build and deploy applications. The quality of service, security, performance, energy consumption, among others, are essential aspects of their deployment, management, and orchestration. Inappropriate resource allocation can lead to resource contention, entailing reduced performance, poor energy efficiency, and other potentially damaging effects. In this paper, we present a set of online job allocation strategies to optimize quality of service, energy savings, and completion time, considering contention for shared on-chip resources. We consider the job allocation as the multilevel dynamic bin-packing problem that provides a lightweight runtime solution that minimizes contention and energy consumption while maximizing utilization. The proposed strategies are based on two and three levels of scheduling policies with container selection, capacity distribution, and contention-aware allocation. The energy model considers joint execution of applications of different types on shared resources generalized by the job concentration paradigm. We provide an experimental analysis of eighty-six scheduling heuristics with scientific workloads of memory and CPU-intensive jobs. The proposed techniques outperform classical solutions in terms of quality of service, energy savings, and completion time by 21.73–43.44%, 44.06–92.11%, and 16.38–24.17%, respectively, leading to a cost-efficient resource allocation for cloud infrastructures.

Performance-Energy Trade-off in Modern CMPs

Chip multiprocessors (CMPs) are ubiquitous in all computing systems ranging from high-end servers to mobile devices. In these systems, energy consumption is a critical design constraint as it constitutes the most significant operating cost for computing clouds. Analogous to this, longer battery life continues to be an essential user concern in mobile devices. To optimize on power consumption, modern processors are designed with Dynamic Voltage and Frequency Scaling (DVFS) support at the individual core as well as the uncore level. This allows fine-grained control of performance and energy. For an n core processor with m core and uncore frequency choices, the total DVFS configuration space is now m (n+1) (with the uncore accounting for the + 1). In addition to that, in CMPs, the performance-energy trade-off due to core/uncore frequency scaling concerning a single application cannot be determined independently as cores share critical resources like the last level cache (LLC) and the memory. Thus, unlike the uni-processor environment, the energy consumption of an application running on a CMP depends not only on its characteristics but also on those of its co-runners (applications running on other cores). The key objective of our work is to select a suitable core and uncore frequency that minimizes power consumption while limiting application performance degradation within certain pre-defined limits (can be termed as QoS requirements). The key contribution of our work is a learning-based model that is able to capture the interference due to shared cache, bus bandwidth, and memory bandwidth between applications running on multiple cores and predict near-optimal frequencies for core and uncore.

Extending Performance-Energy Trade-offs Via Dynamic Core Scaling

Modern processors often need to switch among different power states based on usage scenarios, energy availability, and thermal conditions. Dynamic voltage and frequency scaling (DVFS) is a commonly used power management strategy to trade off performance and energy. As transistor scaling is reaching its limit, the viable supply voltage range where DVFS can operate is shrinking, which limits its effectiveness. To extend the performance-energy trade-off capabilities in modern processors, this article proposes dynamic core scaling (DCS) that does not rely on voltage scaling. DCS dynamically adjusts the active superscalar datapath resources so that programs run at a given percentage of their maximum speed while minimizing energy consumption at the same time. Since DCS does not need voltage scaling, it can be combined with DVFS to achieve greater energy savings. To effectively manage performance-energy trade-offs using a combination of DCS and DVFS, this article proposes an oracle controller that demonstrates the optimal control strategy, and two practical controllers that are applicable in real implementations. Evaluations using an 8-way superscalar processor implemented in a 45-nm circuit show that DCS is more effective in performance-energy trade-offs than DVFS at the high performance end for a number of SPEC CPU2000 benchmarks. When used together with DVFS, DCS saves an additional 20 percent of a full-size core’s energy on average. At the minimum operating voltage, DVFS hits its limit, while DCS is still able to achieve an average of 46 percent further energy reduction.

Processor Energy–Performance Range Extension Beyond Voltage Scaling via Drop-In Methodologies

This work introduces reconfiguration for energy–performance adaptation beyond conventional voltage scaling in microcontroller-based systems. Coordinated thread-level processor and row-level memory reconfiguration are enabled by an architecture-agnostic methodology. The latter requires low design and integration effort while reusing existing macros, including third-party intellectual properties (IPs) in an obfuscated or encrypted form. The methodology represents a drop-in solution that is applicable to extend the energy–performance tradeoff in existing designs. The proposed approach was demonstrated with a testchip implementing an ARM Cortex-M0 processor with 16-KB SRAM in 40 nm. The system demonstrates 1.8X throughput boost at nominal voltage (1.1 V) and 1.3X energy reduction at the minimum energy point (0.51 V) compared to voltage scaling with no reconfiguration. Such energy–performance range extension is achieved at 10.3% area overhead, which is mostly due to processor reconfiguration. Overall, the proposed approach reduces energy in the common case and increases performance when demanded, surpassing the capabilities of conventional voltage scaling, while reusing existing IPs and retaining the original software stack.

Joint Computation Offloading and Scheduling Optimization of IoT Applications in Fog Networks

In recent times, fog computing becomes an emerging technology that can exhilarate the cloud services towards the network edge for increasing the speeds up of various Internet-of-Things (IoT) applications. In this context, integrating priority-aware scheduling and data offloading allow the service providers to efficiently handle a large number of real-time IoT applications and enhance the capability of the fog networks. But the energy consumption has become skyrocketing, and it gravely affects the performance of the fog networks. To address this issue, in this paper, we introduce an Energy-Efficient Task Offloading ( EETO ) policy combined with a hierarchical fog network for handling energy-performance trade-off by jointly scheduling and offloading the real-time IoT applications. To achieve this objective, we formulate a heuristic technique for assigning a priority on each incoming task and formulate a stochastic-aware data offloading issue with an efficient virtual queue stability approach, namely the Lyapunov optimization technique. The proposed technique utilizes the current state information for minimizing the queue waiting time and overall energy consumption while meeting drift-plus-penalty . Furthermore, a constraint restricted progressive online task offloading policy is incurred to mitigate the backlog tasks of the queues. Extensive simulation with various Quality-of-Service (QoS) parameters signifies that the proposed EETO mechanism performs better and saves about 23.79% of the energy usage as compared to the existing ones.

An Adaptive CPU-GPU Governing Framework for Mobile Games on big.LITTLE Architectures

Games have been one of the most popular applications on smartphones. In order to meet the increasing computational complexity of mobile games, smartphones are now equipped with heterogeneous CPU multi-core architectures like big.LITTLE as well as high-performance GPUs. However, the integrated CPUs and GPUs drain the battery quickly, which has become a bottleneck for improving user experience. In addition to traditional Dynamic Voltage and Frequency Scaling (DVFS) technique for CPUs and GPUs power reduction, heterogeneous multi-core processors, such as the big.LITTLE architecture, have been designed to offer more opportunity for performance-energy tradeoffs. But current processor governors in smartphones can not exploit these power-saving mechanisms wisely, causing considerable energy waste. In this article, we propose a CPU-GPU governing framework that recognizes the performance demand for different game scenes, and select the most energy-efficient hardware configuration for the corresponding scenes. We implement our framework on an ODROID-XU4 mobile platform, and the experiments show that our framework can achieve 26.7, 16.6, and 10.5 percent power saving on average without compromising user experience when compared to the default governor used in our platform and two governors proposed by other researchers, respectively.

Smartphone processor architecture, operations, and functions: current state-of-the-art and future outlook: energy performance trade-off

Balancing energy–performance trade-offs for smartphone processor operations is undergoing intense research considering the challenges with the evolving technology of mobile computing. However, to guarantee energy-efficient processor operation, layout, and architecture, it is necessary to identify and integrate optimization techniques and parameters influencing energy–performance trade-off in various processor activity domains. Existing literature on energy optimization in smartphones focuses primarily on individual sub-domains such as OS, GPU, and cloud offloading methods. It reflects multiple smartphone processor activities domains as being the most discussed but less integrated. Through this study, we intend to provide the current state-of-the-art energy optimization techniques for smartphone processor operations. It also classifies multiple energy-draining processor operations along with their thorough discussion of methodologies and popular optimization techniques. The study models smartphone processor sub-components highlighting conventional techniques and performance parameters among its varied domains affecting the device’s energy performance along with significant energy drain minimization without any serious performance degradation. The study analyzes these approaches in the context of applicability, performance, and expected future demands along with revealing limitations of those approaches and open research issues prevailing in the available literature. Finally, we conclude our study by summarizing the current state of the art for smartphone processor activities power consumption.

Energy-performance management in battery powered reconfigurable processors for standalone IoT systems

Reconfigurable processors are gaining more attention in battery powered portable platforms like IoTs. As the FPGA based reconfigurable processors are power hungry and the energy source is limited in battery powered system, energy management is vital for such systems. This paper addresses an energy-performance management method in battery powered reconfigurable processors for standalone IoT systems. It considers battery characteristics and nonlinearities to select the best set of task types with respect to energy-performance management policies while satisfying area and energy consumption constraints in different situations. The proposed method decides based on battery level and rate of discharge to adopt two different strategies: (1) energy-performance trade-off, (2) energy minimization. The proposed method benefits from three tools to reduce the power consumption: (1) reducing the total number of performed reconfiguration during the battery life, (2) using battery-ware scheduling, (3) reducing occupied resources. The results show that the energy-performance management system extends battery lifetime by 34% in energy minimizing scenario. Moreover, it boosts the average performance by 31% while extending the battery lifetime by 25% in the complete discharging scenario. In the situation that the battery is charging, the energy-performance management system has achieved an increment of 74% in performance compared to (3).

Analysis of a variable service speed single server queue with batch arrivals and general setup time

In this paper, we consider an MX/M/1/GSET-VARI queue, which is a single server queue with a variable service speed, batch arrivals and general setup times. The service speed of the server is proportional to the number of jobs in the system. Our model is motivated by power-aware servers in data centers where dynamic scaling techniques are used. In this paper, we obtain the necessary and sufficient condition for the stability of the system and derive an expression for the probability generating function of the number of jobs in the system. In addition, we present some numerical results in order to show the energy-performance trade-off of our queueing system.

The $$CiS^2$$: a new metric for performance and energy trade-off in consolidated servers

The increased use of cloud services has turned the virtualization in the main technology that supports cloud datacentres. To reduce the increment of power consumption caused in datacenters due to the addition of physical servers, system administrators are using virtual machine consolidation (VMC) techniques which tries to allocate the adequate number of virtual machines per physical server. Therefore, VMC increases server resources utilization and as a consequence its performance degradation and the energy consumption too. Then, a trade-off between the performance and the energy consumption exists when consolidating virtual machines. This trade-off is difficult to quantify and also to determine the servers efficiency taking into account a specific number of allocated virtual machines. Because of this, it is crucial for system administrators having a simple metric that assists the VMC making-decision process. In this paper, we propose the $$CiS^2$$ index, a metric to quantify this performance-energy trade-off. Also, this index can help system administrators to decide about the servers’ efficiency through benchmarking and to select the most efficient server through a proposed algorithm. Besides, we propose a simple graphical representation of the index to distinguish graphically the efficient and non-efficient server consolidations. We validate the index in a theoretical manner and performing real experiments in different physical servers under CPU workload saturation. Obtained results show that the proposed index reflects the performance-energy trade-off behaviour and it helps systems’ administrators when consolidating virtual machines.

An energy efficient FPGA partial reconfiguration based micro-architectural technique for IoT applications

Low power consumption and high computational performance are two important processor design goals for IoT applications. Achieving both design goals in one processor architecture is challenging due to their conflicting requirements. This paper introduces a reconfigurable micro-architectural level technique that allows a Reduced Instruction Set Computing (RISC) processor to support IoT applications with different performance and energy trade-off requirements. The processor can be reconfigured into either multi-cycle execution mode (low computational speed with low dynamic power consumption) or pipeline execution mode (high computational speed at the expense of high dynamic power), based on dynamic workload characteristics in IoT applications. Switching between modes is accomplished by exploiting the partial reconfiguration (PR) feature offered by the recent advancements in modern FPGAs. A RISC processor was designed based on the proposed micro-architectural level technique and implemented on FPGA as IoT sensor node. Experimental results demonstrate that the proposed technique with reconfigurable micro-architecture is able to significantly reduce the dynamic energy consumption, compared to conventional multi-cycle and pipeline only micro-architectures, while allowing better performance-energy trade-off in IoT applications.

Efficient virtual machine placement algorithms for consolidation in cloud data centers

Dynamic Virtual Machine (VM) consolidation is a successful approach to improve the energy efficiency and the resource utilization in cloud environments. Consequently, optimizing the online energy-performance tradeoff directly influences quality of service. In this study, algorithms named as CPU Priority based Best-Fit Decreasing (CPBFD) and Dynamic CPU Priority based Best-Fit Decreasing (DCPBFD) are proposed for VM placement. A number of VM placement algorithms are implemented and compared with the proposed algorithms. The algorithms are evaluated through simulations with real-world workload traces and it is shown that the proposed algorithms outperform the known algorithms. The simulation results clearly show that CPBFD and DCPBFD provide the least service level agreement violations, least VM migrations, and efficient energy consumption.

On energy-aware M/G/1-LAS queue with batch arrivals

We analyze an energy-aware MX/G/1 queue under LAS scheduling with a setup delay and an idle timer that controls the delay before the server enters a sleep state. Through a classical busy period analysis, the expression for the mean conditional delay is derived, which generalizes the earlier well-known result for the ordinary M/G/1-LAS queue. We also analyze the performance-energy tradeoff and show that two well-known cost metrics, weighted sum and product of the mean delay and mean power, are minimized by setting the timer equal to zero or infinite, i.e., a finite idle timer is never used.

Near-optimal dispatching policy for energy-aware server clusters

A server cluster can be modeled as a set of parallel queues, and the dispatcher decides to which queue the arriving jobs are routed. We consider an energy-aware dispatching system in a Markovian setting, where each server, upon becoming empty, enters a sleep mode to save energy, and to activate the server after sleep incurs an additional setup delay cost. We seek to optimize the performance-energy trade-off by applying the so-called Whittle index approach. As our main result, we derive sufficient conditions for the system parameters under which the problem is provably indexable, and also determine the corresponding Whittle index values explicitly. Our numerical experiments demonstrate that the resulting energy-aware Whittle index policy is able to perform very close to the numerically solved optimal policy and outperforms all considered reference policies.