Increase In Execution Time Research Articles

Ensuring confidentiality and privacy of cloud data using a non-deterministic cryptographic scheme.

The amount of data generated by electronic systems through e-commerce, social networks, and data computation has risen. However, the security of data has always been a challenge. The problem is not with the quantity of data but how to secure the data by ensuring its confidentiality and privacy. Though there are several research on cloud data security, this study proposes a security scheme with the lowest execution time. The approach employs a non-linear time complexity to achieve data confidentiality and privacy. A symmetric algorithm dubbed the Non-Deterministic Cryptographic Scheme (NCS) is proposed to address the increased execution time of existing cryptographic schemes. NCS has linear time complexity with a low and unpredicted trend of execution times. It achieves confidentiality and privacy of data on the cloud by converting the plaintext into Ciphertext with a small number of iterations thereby decreasing the execution time but with high security. The algorithm is based on Good Prime Numbers, Linear Congruential Generator (LGC), Sliding Window Algorithm (SWA), and XOR gate. For the implementation in C#, thirty different execution times were performed and their average was taken. A comparative analysis of the NCS was performed against AES, DES, and RSA algorithms based on key sizes of 128kb, 256kb, and 512kb using the dataset from Kaggle. The results showed the proposed NCS execution times were lower in comparison to AES, which had better execution time than DES with RSA having the longest. Contrary, to existing knowledge that execution time is relative to data size, the results obtained from the experiment indicated otherwise for the proposed NCS algorithm. With data sizes of 128kb, 256kb, and 512kb, the execution times in milliseconds were 38, 711, and 378 respectively. This validates the NCS as a Non-Deterministic Cryptographic Algorithm. The study findings hence are in support of the argument that data size does not determine the execution time of a cryptographic algorithm but rather the size of the security key.

A Novel Algorithm for Optimal Harmonic Load Flow including Harmonic Compensation

Load flow analysis is widely used for finding voltage at various parts of a large, interconnected power system. The exponential increase in the use of power electronic devices and the noticeable percentage of integration of renewable energy sources in modern power systems result in the occurrence of non-sinusoidal voltages at various parts of the power system. By reformation of the Newton-Raphson load flow method, a few harmonic load flow algorithms have been developed for modern transmission systems. However, for larger transmission systems, complexity and heavy computational burden are often encountered in the load flow solution process due to the increased size of the Jacobean matrix that further increases with multiple non-linear load buses and compensation requirements. This leads to convergence problems and increased execution time. In this paper, an optimal harmonic load flow algorithm is proposed, that employs a modified PSO technique to select control variables and the harmonic load flow method to find solutions for load flow analysis with a reduced size of the Jacobean matrix. The formulated method uses a simple technique to take into account nonlinear loads and results in faster convergence. The novelty of the algorithm is that an optimal load flow solution with the desired amount of reactive and harmonic compensation currents is obtained. The algorithm can help in getting load flow solutions in future power systems with distorted voltage and currents with compensation. The performance of the proposed algorithm is tested on a modified IEEE 30-bus system with multiple non-linear load buses and is validated with the Simulink model of the system.

Methodology and Performance Assessment of Three-Dimensional Vehicular Ad-Hoc Network Simulation

Packet-based simulation is a key tool for the research and development of Vehicular Ad-hoc Networks (VANETs). Over the last decade, many models throughout the communication stack have been presented, which have increased the degree of realism that can be achieved with popular simulation frameworks. Nevertheless, the three-dimensional aspects of many real-world traffic situations barely find consideration. In this paper, we present a holistic approach to simulate large-scale three-dimensional VANET scenarios. We briefly summarize our previously presented models covering different aspects of communication in 3D scenarios, including an environmental diffraction model, an n-ray ground interference model, and the consideration of multi-floor communication. We then describe the principle of a model selector, which applies the appropriate models depending on the environment of the currently transmitted packet. Subsequently, we use the outlined methodology implemented in our Veins 3D framework to simulate a large urban reference scenario. The results differ significantly from comparable 2D simulations, demonstrating the necessity of three-dimensional considerations. However, they also show strongly increased execution times. Therefore, we further suggest different approaches to improve the simulation performance. Based on these optimizations, simulation durations in the same order of magnitude as a comparable 2D simulation can be achieved.

FITEVT: A FORTRAN program for arrival-time analysis of nuclear-decay events

FITEVT is a FORTRAN program designed to analyze recorded arrival times of nuclear-decay events affected by dead-time. The primary goal of the program is precise and accurate determination of half-lives. Its method involves imposition of a known sufficiently long extending dead-time to the recorded event sequence, so that the original dead-time effects are completely obliterated and the remaining live-times of the survived events are known exactly. Upon completion of the analysis, the arrival-time spectrum and the live-time spectrum are predicted and compared to those constructed from the survived events. Program summaryProgram title: FITEVTLicensing provisions: MITCPC Library link to program files:https://doi.org/10.17632/scg6p9jxjk.1Programming language: FORTRAN (using double precision)Nature of problem: Accurate determination of the ideal event rate from the observed event rate in a nuclear decay can be a challenging task due to the inevitable presence of detection system's dead-time, whose nature and extent are typically unknown. Consequently, the results of nuclear half-life measurements may depend on the method used to correct for event losses due to dead-time. Therefore, there is a need for an exact method of data analysis and the means to apply it in order to ensure reproducibility, accuracy and optimized precision of the results.Solution method: FITEVT performs arrival-time analysis of the recorded events, in which a known, sufficiently large extending dead-time is imposed by means of software, so that in the resulting set of survived events the original dead-time effects are completely replaced by the effects of the imposed dead-time. As a result, the dead-time following each survived event and the remaining live-time preceding each surviving event are known exactly. This allows for application of the maximum-likelihood method of estimation, in which the half-lives as well as the other parameters of the ideal-decay-rate function can be accurately determined.Additional comments including restrictions and unusual features: In its original form the program can handle contributions from a constant event rate (i.e., a typical background) plus up to 4 different exponentially-decaying event rates. One of these 4 contributions is allowed to be negative, so that the program can be applied to the general case involving a single two-component decay chain. The employed method of analysis produces accurate results even when the product of the ideal event rate (ρ) and the extending dead-time parameter (τe) is as large as 51.2 [1]. Example used in the present paper involves simulated events having initial rate of 105 s−1, to which extending dead-time of 64 μs is imposed.Data analysis is based on maximum-likelihood principle, in which minimization of the deviance involves calculations in double precision and determination of its derivatives is based on analytical expressions. This is done in order to optimize precision and accuracy of the results as well as the program execution time.With the provided input data and a typical contemporary laptop computer, the program execution typically lasts about 5 minutes. Reducing the imposed dead-time to zero increases the number of survived events approximately by a factor of 6, but the program execution time increases by about a factor of 3.

A Layer & Request Priority-based Framework for Dynamic Resource Allocation in Cloud- Fog - Edge Hybrid Computing Environment

One of the most promising frameworks is the fog computing paradigm for time-sensitive applications such as IoT (Internet of Things). Though it is an extended type of computing paradigm, which is mainly used to support cloud computing for executing deadline-based user requirements in IoT applications. However, there are certain challenges related to the hybrid IoT -cloud environment such as poor latency, increased execution time, computational burden and overload on the computing nodes. This paper offers A Layer & Request priority-based framework for Dynamic Resource Allocation Method (LP-DRAM), a new approach based on layer priority for ensuring effective resource allocation in a fog-cloud architecture. By performing load balancing across the computer nodes, the suggested method achieves an effective resource allocation. Unlike conventional resource allocation techniques, the proposed work assumes that the node type and the location are not fixed. The tasks are allocated based on two constrain, duration and layer priority basis i.e, the tasks are initially assigned to edge computing nodes and based on the resource availability in edge nodes, the tasks are further allocated to fog and cloud computing nodes. The proposed approach's performance was analyzed by comparing it to existing methodologies such as First Fit (FF), Best Fit (BF), First Fit Decreasing (FFD), Best Fit Decreasing (BFD), and DRAM techniques to validate the effectiveness of the proposed LP-DRAM.

DEPO: A dynamic energy‐performance optimizer tool for automatic power capping for energy efficient high‐performance computing

AbstractIn the article we propose an automatic power capping software tool DEPO that allows one to perform runtime optimization of performance and energy related metrics. For an assumed application model with an initialization phase followed by a running phase with uniform compute and memory intensity, the tool performs automatic tuning engaging one of the two exploration algorithms—linear search (LS) and golden section search (GSS), finds a power cap optimizing a given metric and sets it for the remaining computations. The considered metrics include energy (E), energy‐delay sum, energy‐delay product. We present experimental results obtained for a set of benchmarks that differ in compute and memory intensity—parallel custom built OpenMP implementations of: numerical integration, heat distribution simulation (HEAT), fast Fourier transform (FFT), and additionally NAS parallel benchmarks: CG, MG, BT, SP, and LU. Tests were performed using multi‐core CPUs that are representatives of modern servers and the desktop family: 2 Intel Xeon E5‐2670 v3 CPU (Haswell‐EP) and Intel i7‐9700K CPU (Coffee Lake). The results show that our approach enabled considerable improvements for the tested metrics, for example, for HEAT and Coffee Lake we minimized energy by 50% at the cost of a 15% increase in execution time (LS), for FFT energy was minimized by 40% at a 25.5% increase in execution time (GSS), for SP and Haswell energy was minimized by 25% at the cost of an 18.5% time increase and for Coffee Lake energy was decreased by 56% with a 12% time increase.

The Central Medial Thalamic Nucleus Facilitates Bilateral Movement Execution in Rats

Intralaminar thalamic nuclei, including the central medial nucleus (CMT), have been classically implicated in the control of attentional functional states such as sleep-wake transitions. In rodents, the CMT innervates large cortical and subcortical areas bilaterally, including sensorimotor regions of the cortex and striatum, but its contribution to motor function, which regularly develops in faster temporal scales than attentional states, is still far from being completely understood. Here, by using a novel behavioral protocol to evaluate bilateral coordination in rats, combined with electrophysiological recordings and optogenetic manipulations, we studied the contribution of the CMT to motor control and coordination. We found that optogenetic stimulation of the central region of the CMT produced bilateral recruitment of neural activity in the sensorimotor cortex and striatum. The same type of stimulations produced a significant increase in bilateral movement coordination of the forelimbs accompanied by a decrease in movement trajectory variability. Optogenetic inactivation of the CMT did not affect motor execution but significantly increased execution times, suggesting less interest in the task. Altogether, our results indicate that brief CMT activations create windows of synchronized bilateral cortico-striatal activity, suitable to facilitate motor coordination in temporal scales relevant for motor execution.

Rational or emotional? Prohibitive voice of outsourced employees in a time trajectory perspective

Does outsourced employees' risk behavior depend more on rationality or emotion in temporary interorganizational project‐based teams? Combining trust‐related research and a time trajectory perspective, this study re‐examines the relationship between trust and outsourced employees' prohibitive voice in interorganizational project‐based teams. Two‐wave survey data were collected from 286 outsourced employees and their supervisors across 52 interorganizational teams in China. Empirical results show that outsourced employees' prohibitive voice depends more on cognition‐based trust than on affect‐based trust. With project execution time increases, the promoting effect of affect‐based trust on prohibitive voice shows an increase, while the impact of cognition‐based trust demonstrated little variation. Moreover, outsourced employees' perceptions of leader‐member exchange (LMX) differentiation mediates the moderating effect of project execution time on the relationship between affect‐based trust and prohibitive voice. However, the mediated moderation effect is not significant for cognition‐based trust. Theoretical and practical implications for project‐based team management are discussed.

Energy-efficient IoT routing based on a new optimizer

Several difficulties are generally encountered when solving many-objective problems (fitted with three or more conflictual objectives) by applying multi-objective algorithms (resolving two or three objectives), especially those related to their performances. These issues are, in fact, related to their increased execution time, to the strength of recombination/mutation variation or even to the degree of convergence and diversity of the solution. In this study, the indoor routing of IoT devices is investigated and resolved using a new hybrid ant-genetic algorithm. The latter is assessed using numerical and statistical tests. Then, simulation and experimental real prototyping are achieved to show the efficiency of the introduced algorithm compared with that of the existing methods.

POS1005 ACTIVE INFLAMMATORY LESIONS OF THE SPINE IN SACROILIAC JOINTS MRI (SIJs-MRI): AN OVERLOOKED FINDING BY RADIOLOGISTS.

BackgroundEstablishing a diagnosis for axial spondylarthritis (axSpA) is based on the combined presence of several clinical and laboratory features in association with imaging: the presence of active inflammatory lesions (bone marrow edema, BME) in sacroiliac joints (SIJs), also associated with structural lesions, is the main element in favor of a diagnosis of axSpA and SIJ MRI is currently the main recommended technique for detecting inflammation, enabling early diagnosis in patients with normal radiographic findings. MRI of the spine is generally not recommended for axSpA diagnosis, since spine involvement is considered less sensitive and specific. Several studies explored the incremental value of spine MRI for diagnosing patients with axSpA, especially when the findings of SI are equivocal, and “predicting” clinical disease activity.ObjectivesTo evaluate the effectiveness of the MRI study dedicated to SIJs to identify the inflammatory involvement of the lumbar spine in patients with active sacroiliitis and clinical diagnosis of axSpA.MethodsMRI of 96 patients (26 M, 70 F, mean age 48 years) with SIJs-MRI positive for inflammatory involvement according to ASAS criteria, were evaluated retrospectively. The presence of signs of inflammation of posterior spinal elements and vertebral bodies included in the field of vision (FOV) was therefore researchedResultsOf 96 patients with active sacroiliitis, 88 had concomitant inflammatory lesions of the lumbar spine. Facet capsulitis (87.6 %) and posterior enthesitis (42.2 %) were the most common lesions and these lesions frequently coexisted (39.3%). Twelve patients with facet capsulitis presented also facet arthritis. Only 20 out of 96 patients (22%) had vertebral bodies inflammatory lesions: 15 corner inflammatory lesions and 5 aseptical spondylodiscitis.ConclusionOn the basis of current working consensus MRI examination of the spine is not recommended for axSpA diagnosis, because it is considered less sensitive and specific. However the high prevalence (91.7%) of spinal inflammatory changes in patients with active sacroiliitis suggests that these features can be used as an additional element in favor of diagnosis, especially in doubtful cases. This evaluation takes little reporting time, without any increase in MRI execution time.

A Simulation Model of Power Demand Management by Manufacturing Enterprises under the Conditions of Energy Sector Transformation

This paper addresses electricity consumption management in manufacturing enterprises. The research aims to provide manufacturing enterprises with an effective tool to control electricity costs. Recently, some factors have been observed to affect the rapid changes in the operating conditions of enterprises. These include the transformation of the power sector toward renewable energy, the disruption of supply chains resulting from a coronavirus pandemic, political crises, and process automation. A method for the analysis and management of electricity consumption in enterprises based on simulation modeling is proposed. The simulation model contains predefined objects representing physical system elements and the data processing algorithm. The production order execution time, energy consumption, employee overtime, and machine load are included in the model. The results show that it is possible to determine the level of power available for the process completion and its influence on the production volume and realization time. In the studied case, when the available power was reduced by half, there was an increase in order execution time of nearly 25 percent and an increase in energy consumption of nearly 15 percent. The method can be used in the operational activities of enterprises as well as extended to different types of production processes.

Toward Industrial Private AI: A Two-Tier Framework for Data and Model Security

With the advances in 5G and IoT devices, industries are vastly adopting artificial intelligence (AI) techniques for improving classification and prediction-based services. However, the use of AI also raises concerns regarding data privacy and security that can be misused or leaked. Private AI was recently coined to address the data security issue by combining AI with encryption techniques, but existing studies have shown that model inversion attacks can be used to reverse engineer the images from model parameters. In this regard, we propose a federated learning and encryption-based private (FLEP) AI framework that provides two-tier security for data and model parameters in an Industrial IoT environment. We propose a three-layer encryption method for data security and provided a hypothetical method to secure the model parameters. Experimental results show that the proposed method achieves better encryption quality at the expense of slightly increased execution time. We also highlight several open issues and challenges regarding the FLEP AI framework's realization.

Multi-Channel CNN based image classification using SKIP connection and MSVM

Deep learning innovations have paved way for effective classification algorithms using the Convolutional Neural Networks (CNNs). The current scenario uses very deep networks to improve the overall efficiency. This deep nature will result in increased complexity, a high number of parameters, increased execution time, and a more complex hardware platform for execution. Our research focuses on minimizing this complex nature of architecture. To achieve this, we employed the multi-channel CNN with a shallow layers approach, which consists of the main channel and side channels. The proposed work uses the Multi class Support Vector Machione (MSVM) as classifier and three distinct architectures with varied filter widths to acquire different performance characteristics. All these models are trained and tested on a brain tumor type database and performance parameters are compared to deep architectures like the Alexnet, VGG16, VGG19, and Resnet 50. When compared to deep architectures for the same database, our model can reduce the overall number of parameters and execution time with comparable accuracy. To improve the overall efficiency, our final architecture includes a skip connection.

RESEARCH ON FUZZY SCHEDULING OF CLOUD COMPUTING TASKS BASED ON HYBRID SEARCH ALGORITHMS AND DIFFERENTIAL EVOLUTION

In the process of task scheduling, due to the large search space of resources, it takes a long time to allocate appropriate resources for tasks, which results in the increase of execution time and execution cost of algorithms. For this reason, this paper proposes a cloud computing task fuzzy scheduling strategy based on hybrid search algorithm and differential evolution, which incorporates the classification based on normal distribution and a variety of mutation strategies on the basis of standard differential evolution algorithm. In mutation strategy, the individual difference vector assigns priority to each task and arranges tasks according to resource allocation rules. It improves the slow convergence speed and easy to fall into local optimum of standard difference algorithm, and can effectively solve the task scheduling problem of cloud computing. A cloud computing task scheduling algorithm with time and cost constraints is designed and tested in a simulation environment. The experimental results show that the algorithm can not only shorten the task processing time and reduce the execution cost, but also fully satisfy the users’ actual quality of service requirements by adjusting the weights of time and cost factors.

A ML-based resource utilization OpenCL GPU-kernel fusion model

Massive data parallelism can be achieved by using general-purpose graphics processing units (GPGPU) with the help of the OpenCL framework. When smaller data with higher GPU memory is executed, it results in a low resource utilization ratio and energy inefficiencies. Up until now, there is no existing model to share GPU for further execution. In addition, if the kernel pair requires the same computation resource, then kernel merging also results in a significant increase in execution time. Therefore, optimal device selection, as well as kernel merging, can significantly speed up the execution performance for a batch of jobs. This paper proposes a kernel merging method that leads to high GPU occupancy. As a result, it reduces execution time and increases GPU utilization. Additionally, a machine learning (ML)-based GPU sharing mechanism is presented to select pairs of kernels in OpenCL frameworks. The model first selects suitable architecture for the jobs and then merges GPU kernels for better resource utilization. From all the GPU candidates, the optimal pair of the kernel concerning data size is selected. The experimental results show that the developed model can achieve 0.91 F1-measure for device selection and 0.98 for the scheduling scheme of kernel merging.

Knowledge Source Rankings for Semi-Supervised Topic Modeling

Recent work suggests knowledge sources can be added into the topic modeling process to label topics and improve topic discovery. The knowledge sources typically consist of a collection of human-constructed articles, each describing a topic (article-topic) for an entire domain. However, these semisupervised topic models assume a corpus to contain topics on only a subset of a domain. Therefore, during inference, the model must consider which article-topics were theoretically used to generate the corpus. Since the knowledge sources tend to be quite large, the many article-topics considered slow down the inference process. The increase in execution time is significant, with knowledge source input greater than 103 becoming unfeasible for use in topic modeling. To increase the applicability of semisupervised topic models, approaches are needed to speed up the overall execution time. This paper presents a way of ranking knowledge source topics to satisfy the above goal. Our approach utilizes a knowledge source ranking, based on the PageRank algorithm, to determine the importance of an article-topic. By applying our ranking technique we can eliminate low scoring article-topics before inference, speeding up the overall process. Remarkably, this ranking technique can also improve perplexity and interpretability. Results show our approach to outperform baseline methods and significantly aid semisupervised topic models. In our evaluation, knowledge source rankings yield a 44% increase in topic retrieval f-score, a 42.6% increase in inter-inference topic elimination, a 64% increase in perplexity, a 30% increase in token assignment accuracy, a 20% increase in topic composition interpretability, and a 5% increase in document assignment interpretability over baseline methods.

A Unified Mechanism for Cloud Scheduling of Scientific Workflows

Scheduling plays a vital role in the efficient utilization of the available resources in clouds. This paper investigates the capabilities of the current scheduling algorithms of WorkFlowSim framework for processing scientific workflows. These investigations used four different sizes of workloads each, for, five well-known workflows. It was revealed that none of the existing algorithms is capable of efficiently executing all the four sizes of workload for the complete set of workflows. Different algorithms performed better, when they were applied to various workloads of a particular workflow. This fact was used in developing an improved unified mechanism, which is capable of using an existing algorithm that performed well in the past, against the given workload. Evaluation results showed that the proposed mechanism improved over the existing algorithms for 4 out of 5 workflows (Epigenomics, Inspiral, Cyber Shake, and Montage), when tested against an aggregated load of all sizes, in terms of simulation time. For the workflow named SIPHT, however, it responded exactly the same as Max-Min algorithm. The minimum and maximum improvements, against the existing best and worst algorithms, in percentage, for Epigenomics, Inspiral, SIPHT, Cyber Shake and Montage were 16-63, 30-68, 0-69, 30-68, and 9-71 in corresponding order. This work has an additional overhead in terms of a dedicated module to find and store algorithmic performance. It is, however, required once and, thus, the increase in execution time might be marginal. The future work intends to check the impact of compute time towards optimization parameters such as makespan, pricing and deadlines.

Clustering Big Data Based on Distributed Fuzzy K-Medoids: An Application to Geospatial Informatics

The advent of big data related to spatial position knowledge, called geospatial big data, provides us with opportunities to recognize the urban environment. Existing database processing methods are inadequate to rapidly provide reliable results in a geospatial big data context due to the need for defining approximation “measures” and the increasing execution time for the queries. The clustering method yields the functional effects. How to scale and accelerate clustering algorithms while maintaining high clustering efficiency, on the other hand, remains a significant challenge. The paper’s primary contribution is the introduction of a modified hierarchical distributed k-medoid clustering method that is specific to spatial query analysis for big data. To improve the efficiency of the k-medoid algorithm and obtain more precise clusters, the suggested model utilizes the Fuzzy k-Medoids method to overcome outliers in the spatial data set and to deal with data uncertainty. The method is complex in nature since it is not predicated on the number of right clusters. The proposed model is divided into two phases: the first step creates local clusters based on a portion of the entire dataset; this stage makes extensive use of the parallelism paradigm provided by the Apache Spark framework; and the second phase aggregates the local clusters to produce compact and reliable final clusters. The proposed model greatly reduces the amount of knowledge shared during the aggregation process and automatically produces the appropriate number of clusters based on the dataset characteristics. The results show that the proposed model outperforms the traditional K-medoids in terms of accuracy of obtained centers in big data applications.

Dynamic GPU Energy Optimization for Machine Learning Training Workloads

GPUs are widely used to accelerate the training of machine learning workloads. As modern machine learning models become increasingly larger, they require a longer time to train, leading to higher GPU energy consumption. This paper presents GPOEO, an online GPU energy optimization framework for machine learning training workloads. GPOEO dynamically determines the optimal energy configuration by employing novel techniques for online measurement, multi-objective prediction modeling, and search optimization. To characterize the target workload behavior, GPOEO utilizes GPU performance counters. To reduce the performance counter profiling overhead, it uses an analytical model to detect the training iteration change and only collects performance counter data when an iteration shift is detected. GPOEO employs multi-objective models based on gradient boosting and a local search algorithm to find a trade-off between execution time and energy consumption. We evaluate the GPOEO by applying it to 71 machine learning workloads from two AI benchmark suites running on an NVIDIA RTX3080Ti GPU. Compared with the NVIDIA default scheduling strategy, GPOEO delivers a mean energy saving of 16.2% with a modest average execution time increase of 5.1%.

An Energy-Efficient Scheduling Algorithm for Shared Facility Supercomputer Centers

The evolution of high-performance computing is associated with the growth of energy consumption. Performance of cluster computes (is increased via rising in performance and the number of used processors, GPUs, and coprocessors. An increment in the number of computing elements results in significant growth of energy consumption. Power grids limits for supercomputer centers (SCC) are driving the transition to more energy-efficient solutions. Often upgrade of computing resources is done step-by-step, i.e. parts of older supercomputers are removed from service and replaced with newer ones. A single SCC at any time can operate several computing systems with different performance and power consumption. That is why the problem of scheduling parallel programs execution on SCC resources to optimize energy consumption and minimize the increase in execution time (energy-efficient scheduling) is important. The goal of the presented work was the development of a new energy-efficient algorithm for scheduling computing resources at SCC. To reach the goal the authors analyzed methods of scheduling computing resources in a shared facility, including energy consumption minimizing methods. The study made it possible to formulate the problem of energy-efficient scheduling for a set of CCs and propose an algorithm for its solution. Experiments on NPB benchmarks allowed achieving significant reduction in energy consumption with a minor increase of runtime.