Flexible Workload Research Articles

The Work-Related Stress and Well-Being of Teachers—An Exploratory Study within Primary Schools in Italy

This study analyzes the critical aspects of the professional life of primary school teachers in a southern European context. The social changes of the last two decades have profoundly impacted teachers’ well-being, putting a strain on their ability to adapt to sometimes unfair working conditions. For this purpose, we interviewed twenty teachers (M = 38.85; DS = 9.17) from different areas of Italy and analyzed data using the grounded theory method. Through a data analysis, we summarized and emphasized eight core themes. The results show the effects of the gradual impoverishment of the recognition for the role of the teacher, highlighting the difficulty for teachers in renegotiating more flexible workload assessments and reinforcing psycho-emotional skills for developing a teaching methodology that is attentive to the psycho-emotional needs of their pupils. Schools emerge as a rigid working context where a competent organizational function is often absent. The practical psychological implications of this are discussed.

Intra-day optimal power flow considering flexible workload scheduling of IDCs

The online interactive workloads and offline batch computing services of Internet data centers (IDCs) have scheduling flexibility. Batch computing tasks with low latency sensitivity and priority can adjust the processing time and provide flexibility temporally. In contrast, interactive workloads with high latency sensitivity and priority can change the physical server for processing and provide flexibility spatially. Considering the change of power supply cost with time and space, the load flexibility of IDCs brought by Internet workload scheduling can be utilized to respond to the power system for more economical operation. To take full advantage of the flexibility of IDCs’ load demand, IDCs coordinated by an Internet service company (ISC) as a special controllable load with both temporal and spatial flexibility potential are considered. Instead of operating separately in normal optimal power flow (OPF) models, a two-stage interactive framework of the distribution system operator (DSO) and ISC is established. The operating state of the ISC is considered in OPF and is introduced as regularization terms into the objective function, thus fully exploiting its load flexibility. Some operation constraints are also added to guarantee the quality of service (QoS) of the ISC. The effectiveness of the model and the interactive framework are evaluated in a typical 33-bus medium-voltage distribution network. The results validate the further reduction of the power grid operation cost compared with the traditional OPF operation.

Carbon-Aware Computing for Datacenters

The amount of CO <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$_{2}$</tex-math></inline-formula> emitted per kilowatt-hour on an electricity grid varies by time of day and substantially varies by location due to the types of generation. Networked collections of warehouse scale computers, sometimes called Hyperscale Computing, emit more carbon than needed if operated without regard to these variations in carbon intensity. This paper introduces Google’s system for global Carbon-Intelligent Compute Management, which actively minimizes electricity-based carbon footprint and power infrastructure costs by delaying temporally flexible workloads. The core component of the system is a suite of analytical pipelines used to gather the next day’s carbon intensity forecasts, train day-ahead demand prediction models, and use risk-aware optimization to generate the next day’s carbon-aware Virtual Capacity Curves (VCCs) for all datacenter clusters across Google’s fleet. VCCs impose hourly limits on resources available to temporally flexible workloads while preserving overall daily capacity, enabling all such workloads to complete within a day with high probability. Data from Google’s in-production operation shows that VCCs effectively limit hourly capacity when the grid’s energy supply mix is carbon intensive and delay the execution of temporally flexible workloads to “greener” times.

Effect of Stress on Employees’ Job Performance of Commercial Banks in Delta State, Nigeria

The purpose of the study is to determine how stress affects the work performance of employees at selected banks in Delta State, Nigeria. The study, anchored on the Selye stress theory, and adopted the survey research design. The study population consists of all staff of six selected commercial banks in Delta State. A sample size of 276 was used for the study. The study's findings showed that while role ambiguity and organizational change had significant negative effects on employees' job performance, workload and time pressure had a significant positive effect. The study comes to the conclusion that workers' job performance at commercial banks in Delta State is impacted by stress, and recommends amongst others that flexible workload should be incorporated into policies of banks to enhance employee productivity and commitment that will increase corporate survival.

EdgeDR: An Online Mechanism Design for Demand Response in Edge Clouds

The computing frontier is moving from centralized mega datacenters towards distributed cloudlets at the network edge. We argue that cloudlets are well-suited for handling power demand response to help the grid maintain stability due to more flexible workload management attributed to their distributed nature. However, they also require computing demand response to avoid overload and maintain reliability. To this end, we propose a novel online market mechanism, EdgeDR, to achieve cost efficiency in edge demand response programs. At a high level, we observe that the cloudlet operator can dynamically switch on/off entire cloudlets to compensate for the energy reduction required by the power grid or provide enough computing resources to the edge service. We formulate a long-term social cost minimization problem and decompose it into a series of one-round procurement auctions. In each auction instance, we propose to let the cloudlet tenants bid with cost functions of their two-dimension service quality degradation tolerance, and let the cloudlet operator choose the service quality, manage the workload, and schedule the cloudlet activation status. In addition, we present a dynamic payment mechanism for the operator to balance the tradeoff between short-term profit and long-term benefit in more practical scenarios. Via rigorous analysis, we exhibit that our bidding policy is individually rational and truthful; our workload management algorithm has near-optimal performance in each auction; and our overall online algorithm achieves a provable competitive ratio. We further confirm the performance of our mechanism through extensive trace-driven simulations.

AIPerf: Automated machine learning as an AI-HPC benchmark

The plethora of complex Artificial Intelligence (AI) algorithms and available High-Performance Computing (HPC) power stimulates the expeditious development of AI components with heterogeneous designs. Consequently, the need for cross-stack performance benchmarking of AI-HPC systems has rapidly emerged. In particular, the defacto HPC benchmark, LINPACK, cannot reflect the AI computing power and input/output performance without a representative workload. Current popular AI benchmarks, such as MLPerf, have a fixed problem size and therefore limited scalability. To address these issues, we propose an end-to-end benchmark suite utilizing automated machinelearning, which not only represents real AI scenarios, but also is auto-adaptively scalable to various scales ofmachines. We implement the algorithms in a highly parallel and flexible way to ensure the efficiency and optimizationpotential on diverse systems with customizable configurations. We utilize Operations Per Second (OPS), which ismeasured in an analytical and systematic approach, as a major metric to quantify the AI performance. We performevaluations on various systems to ensure the benchmark's stability and scalability, from 4 nodes with 32 NVIDIA Tesla T4 (56.1 Tera-OPS measured) up to 512 nodes with 4096 Huawei Ascend 910 (194.53 Peta-OPS measured), and the results show near-linear weak scalability. With a flexible workload and single metric, AIPerf can easily scaleon and rank AI-HPC, providing a powerful benchmark suite for the coming supercomputing era.

Medicine in Motion: Addressing Physician Burnout Through Fitness, Philanthropy, and Interdisciplinary Community Building

Physician burnout is a public health crisis. Although recent studies underscore its prevalence, there are few rigorous studies examining its prevention, especially among medical students and residents. Prior interventions have centered on mindfulness techniques and flexible workload scheduling, yielding limited success. However, interventions that combine fitness with philanthropy and community building may be more effective. The purpose of this report is 2-fold: first, to provide a review of physician burnout and potential prevention mechanisms and, second, to present a case study of how Medicine in Motion (MiM) addresses these issues. MiM facilitates various athletic workouts, competitions, and other events for students and professionals in medicine, dentistry, nursing, and physical therapy to support wellness and charitable initiatives. This analysis identifies 4 barriers to physicians and those in the health care profession from participating in wellness activities: (1) insufficient awareness, (2) logistical challenges, (3) lack of purpose, and (4) absence of perceived support. To overcome these barriers, MiM provides a model toolkit for starting a grassroots movement against physician burnout that other health care institutions may emulate. Institutions should provide financial support for these wellness programs. Future research is needed to evaluate these combined exercise, philanthropic, and community building efforts.

Exploration of OpenCL Heterogeneous Programming for Porting Solidification Modeling to CPU‐GPU Platforms

SummaryThis article provides a comprehensive study of OpenCL heterogeneous programming for porting applications to CPU–GPU computing platforms, with a real‐life application for the solidification modeling. The aim is to achieve a flexible workload distribution between available CPU–GPU resources and optimize application performance. Considering the solidification application as a use case, we explore the necessary steps required for (i) adaptation of an application to CPU–GPU platforms, and (ii) mapping the application workload onto the OpenCL programming model. The adaptation is based on a reformulation of steps developed previously for CPU–MIC architectures. The mapping process allows us to utilize OpenCL for harnessing CPU and GPU cores using data parallelism, as well as for the management of available compute devices with task parallelism. The resulting OpenCL code's performance and energy efficiency is experimentally studied for two platforms with powerful GPUs of various generations (with Kepler and Volta architectures). The experiments confirm the performance advantage of using computing resources of both GPUs and CPUs. The achieved benefit depends on the relationship between the computing power of CPUs and GPUs. Moreover, this gain entails the growth of the average power that increases the energy consumed during the application execution.

Mitigating Curtailment and Carbon Emissions through Load Migration between Data Centers

Summary As the share of variable renewable energy (VRE) grows in the electric grid, so does the risk of curtailment. While energy storage and hydrogen production have been proposed as solutions to the curtailment problem, they often pose technological and economic challenges. Here, we analyze the potential of data center load migration for mitigating curtailment and greenhouse gas (GHG) emissions. Using historical hourly electricity generation, curtailment, and typical data center server utilization data, we simulate the effect of migrating data center workloads from the fossil-fuel-heavy PJM to the renewable-heavy CAISO. The results show that load migration within the existing data center capacity during the curtailment hours in CAISO has the potential to reduce 113–239 KtCO2e per year of GHG emissions and absorb up to 62% of the total curtailment with negative abatement costs in 2019. Our study demonstrates the overlooked role that data centers can play for VRE integration and GHG emissions mitigation.

Using Ensembles for Accurate Modelling of Manufacturing Processes in an IoT Data-Acquisition Solution

The development of complex real-time platforms for the Internet of Things (IoT) opens up a promising future for the diagnosis and the optimization of machining processes. Many issues have still to be solved before IoT platforms can be profitable for small workshops with very flexible workloads and workflows. The main obstacles refer to sensor implementation, IoT architecture, and data processing, and analysis. In this research, the use of different machine-learning techniques is proposed, for the extraction of different information from an IoT platform connected to a machining center, working under real industrial conditions in a workshop. The aim is to evaluate which algorithmic technique might be the best to build accurate prediction models for one of the main demands of workshops: the optimization of machining processes. This evaluation, completed under real industrial conditions, includes very limited information on the machining workload of the machining center and unbalanced datasets. The strategy is validated for the classification of the state of a machining center, its working mode, and the prediction of the thermal evolution of the main machine-tool motors: the axis motors and the milling head motor. The results show the superiority of the ensembles for both classification problems under analysis and all four regression problems. In particular, Rotation Forest-based ensembles turned out to have the best performance in the experiments for all the metrics under study. The models are accurate enough to provide useful conclusions applicable to current industrial practice, such as improvements in machine programming to avoid cutting conditions that might greatly reduce tool lifetime and damage machine components.

Flexible Multi-Energy Scheduling Scheme for Data Center to Facilitate Wind Power Integration

Due to the intermittent and uncontrollable nature of wind resources and inflexible operation of conventional generation units, they present challenges for the power system to integrate more wind power. With its unique flexibility on the demand side, the data center can be considered as an effective solution to relieve wind curtailment. Moreover, with the help of waste heat recovery module, the data center can reduce the utilization of conventional thermal units especially in the residential heating sector which increases the flexibility of system operation and facilitates more renewable integration. In this paper, a flexible workload management and resource scheduling model are proposed to achieve a multi-energy co-optimization for data center and enhance the integration of wind power. A two-stage stochastic programming model is formulated to address the uncertainties involved in this process. The proposed model is examined by a simulative data center microgrid and the numerical results demonstrate its effectiveness and robustness.

Improved management to limit milk production losses resulting from the transition to once-a-day milking in dairy sheep

There has been increasing interest in once-a-day milking (ODM) in recent years. Despite the loss in milk production due to the reduction in milking frequency, ODM has several advantages for farm management, particularly labour savings. It has already been demonstrated that adapting management can compensate for the reduction in milk production, especially on dairy cow farms based on extensive grazing systems. The objective of this study was therefore to investigate the impact of management adaptations associated with a switch to ODM in a diverse range of dairy ewe farms in the ‘Rayon de Roquefort’, France, at the farm level. For that purpose, a model was developed to evaluate the ability of three flock management adaptations to compensate for losses due to the transition to ODM: (i) increasing the flock size, (ii) increasing the milking period and (iii) increasing grazing in the ewes’ diet. The model results show that depending on the farm, a transition to ODM on the first day of milking reduces the milk yield of the whole flock by 18–19% and causes an 8–16% decrease in household income. When the transition occurs around the turnout date, the impact on the performance of the flock is reduced, but so is the time spent on routine work. In the short term, it is possible to make adaptions to limit losses both in milk production and in household income by lengthening the milking period and/or by making better use of pasture. Conversely, increasing the number of milked ewes does not compensate for the loss of income. For each farm, a scenario was designed to provide a satisfactory trade-off between household income, milk delivery and routine workload. However, although the management adaptations are the same, their modalities of implementation depend on the farm concerned, which have different amounts of leeway. For the diverse range of farms tested, ODM represents a good compromise between flexible workload management and the performance of dairy sheep farms. ODM, during the whole or part of the milking period, is a useful farm management tool to manage both labour and grazing during strategic periods.

Flexible Work Arrangements from Generation and Gender Perspectives: Evidence from Lithuania

The demand for employee-oriented and individualized work arrangements and practices those permit temporal and spatial flexibility in work processes and help to fit changing personal needs and environmental conditions is constantly increasing. A growing number of private companies and public organizations are adopting flexible work arrangements to help employees balance their work–family demands and companies cope with variations in demand, reduce expenses, and be more attractive for employees. Employees representing different age and sex groups have different expectations and priorities towards work, and, likely, they evaluate and respond to flexible work practices also differently. The paper investigates how different genders and generations evaluate and respond to flexibility at work. Empirical research was conducted in Lithuania. Data were collected from a sample of 316 employees (195 women and 121 men). Research revealed a high expression of flexible work arrangements in Lithuanian organizations. Flexible work schedule and flexible work load are dominating in the organizations. Representatives of generations B and Y are most satisfied in the forms of flexible work arrangements applied in their organizations. Hypothesis that women tend to work flexibly more than men was proved partially during the research. Hypothesis that work flexibility in aspect of time and load is more attractive to respondents than flexibility in aspect of working place was totally proved.DOI: http://dx.doi.org/10.5755/j01.ee.29.1.19247

Integrated optimisation of the flexible resource workload balancing and investment project scheduling problem

In this paper, we investigate the extension of the resource investment project scheduling problem: the flexible resource workload balancing and investment project scheduling problem (FRWBIP). The integrated optimisation of FRWBIP is established with the objective of minimising the weighted resource availability cost and workload balancing penalty. A heuristic algorithm is designed to obtain the upper and lower bounds of resource requirements. The skills of the flexible resources are effectively allocated by the minimum cost-maximum flow flexible resource allocation model. A genetic algorithm with an improved serial schedule generation scheme and the resource usage adjustment module is proposed to solve FRWBIP. Every individual of the population is constructed from two lists: the activity priority list and the resource capacity list. Finally, a series of computational experiments are carried out by using the modified standard example library PSPLIB, and the effectiveness of the algorithm is verified by comparison with other literature algorithm.

Integrated optimisation of the flexible resource workload balancing and investment project scheduling problem

Integrated optimisation of the flexible resource workload balancing and investment project scheduling problem

Special Issues on Learning Strategies: Parallels and Contrasts between Australian and Chinese Tertiary Education

Learning strategies are crucial to student learning in higher education. In this paper, there are comparisons of student engagement, feedback mechanism and workload arrangements at some typical universities in Australia and China, which are followed by practical suggestions for active learning. First, an inclusive class would allow learners from different backgrounds to become more engaged in classroom activities. Second, universities should improve feedback mechanisms, making them more timely and helpful to enable students to adapt their learning strategies and allowing teachers to adjust teaching methods to target students effectively. Third, this paper proposes a framework of principles under which the flexible workload of academics should be ensured so that students can learn social skills from administrative staff and have more free time to develop unique thinking and planning capacities.

PanDA for ATLAS distributed computing in the next decade

The Production and Distributed Analysis (PanDA) system has been developed to meet ATLAS production and analysis requirements for a data-driven workload management system capable of operating at the Large Hadron Collider (LHC) data processing scale. Heterogeneous resources used by the ATLAS experiment are distributed worldwide at hundreds of sites, thousands of physicists analyse the data remotely, the volume of processed data is beyond the exabyte scale, dozens of scientific applications are supported, while data processing requires more than a few billion hours of computing usage per year. PanDA performed very well over the last decade including the LHC Run 1 data taking period. However, it was decided to upgrade the whole system concurrently with the LHC’s first long shutdown in order to cope with rapidly changing computing infrastructure. After two years of reengineering efforts, PanDA has embedded capabilities for fully dynamic and flexible workload management. The static batch job paradigm was discarded in favor of a more automated and scalable model. Workloads are dynamically tailored for optimal usage of resources, with the brokerage taking network traffic and forecasts into account. Computing resources are partitioned based on dynamic knowledge of their status and characteristics. The pilot has been re-factored around a plugin structure for easier development and deployment. Bookkeeping is handled with both coarse and fine granularities for efficient utilization of pledged or opportunistic resources. An in-house security mechanism authenticates the pilot and data management services in off-grid environments such as volunteer computing and private local clusters. The PanDA monitor has been extensively optimized for performance and extended with analytics to provide aggregated summaries of the system as well as drill-down to operational details. There are as well many other challenges planned or recently implemented, and adoption by non-LHC experiments such as bioinformatics groups successfully running Paleomix (microbial genome and metagenomes) payload on supercomputers. In this paper we will focus on the new and planned features that are most important to the next decade of distributed computing workload management.

Academic workload: the silent barrier to the implementation of technology-enhanced learning strategies in higher education

The effect of technology-enhanced learning (TEL) strategies in higher education has arguably been transformative despite the not-insignificant barriers existing in this context. Throughout the discourse very little attention has been paid to those primarily responsible for this implementation—academic teaching staff. This paper aims to highlight the impact of academic workload allocations, an often silent barrier to the uptake of TEL strategies in higher education. We will discuss the effects of academic identity and culture, preferential time allocation to associative activities, academic technological capacity, university policies and workload and funding models on the uptake, and implementation on TEL in higher education. Our aim is to highlight the risks to staff, students and institutions should these concerns not be addressed and to propose a model for utilisation by all staff responsible for implementing flexible workload models supportive of further implementation of TEL strategies across the sector.

Dynamic Intelligent Feedback Scheduling in Networked Control Systems

For the networked control system with limited bandwidth and flexible workload, a dynamic intelligent feedback scheduling strategy is proposed. Firstly, a monitor is used to acquire the current available network bandwidth. Then, the new available bandwidth in the next interval is predicted by using LS_SVM approach. At the same time, the dynamic performance indices of all control loops are obtained with a two-dimensional fuzzy logic modulator. Finally, the predicted network bandwidth is dynamically allocated by the bandwidth manager and the priority allocator in terms of the loops' dynamic performance indices. Simulation results show that the sampling periods and priorities of control loops are adjusted timely according to the network workload condition and the dynamic performance of control loops, which make the system running in the optimal state all the time.

Flexible workload generation for HPC cluster efficiency benchmarking

The High Performance Computing (HPC) community is well-accustomed to the general idea of benchmarking. In particular, the TOP500 ranking as well as its foundation--the Linpack benchmark--have shaped the field since the early 1990s. Other benchmarks with a larger workload variety such as SPEC MPI2007 are also well-accepted and often used to compare and rate a system's computational capability. However, in a petascale and soon-to-be exascale computing environment, the power consumption of HPC systems and consequently their energy efficiency have been and continue to be of growing importance, often outrivaling all aspects that focus narrowly on raw compute performance. The Green500 list is the first major attempt to rank the energy efficiency of HPC systems. However, its main weakness is again the focus on a single, highly compute bound algorithm. Moreover, its method of extrapolating a system's power consumption from a single node is inherently error-prone. So far, no benchmark is available that has been developed from ground up with the explicit focus on measuring the energy efficiency of HPC clusters. We therefore introduce such a benchmark that includes transparent energy measurements with professional power analyzers. Our efforts are based on well-established standards (C, POSIX-IO and MPI) to ensure a broad applicability. Our well-defined and comprehensible workloads can be used to, e.g. compare the efficiency of HPC systems or to track the effects of power saving mechanisms that can hardly be understood by running regular applications due to their overwhelming complexity.