Workload Fluctuations Research Articles

Dynamic Dual-Threshold Virtual Machine Merging Method Based on Three-Way Decision

Cloud computing, an emerging computing paradigm, has been widely considered due to its high scalability and availability. An essential stage of cloud computing is the cloud virtual machine migration technology. Nevertheless, the current trigger timing of virtual machine migration in cloud data centers is inaccurate, resulting in insufficient virtual machine consolidation. Furthermore, the high and low workload fluctuations are also a potential symmetrical problem worthy of attention. This paper proposes a virtual machine energy-saving merging method based on a three-way decision (ESMM-3WD). Firstly, we need to calculate the load fluctuation of the physical machine and divide the load fluctuation into three parts. Furthermore, the corresponding mathematical model predicts the load according to the different classification categories. Then, the predicted load value is used to dynamically adjust the threshold to improve the virtual machine merge probability. Finally, the simulation experiment is carried out on the cloud computing simulation platform cloudsim plus. The experimental results show that the virtual machine energy-saving merging method based on the three-way decision proposed in this paper can better reduce the number of migrations, increase the number of physical machines shut down, better improve the probability of virtual machine merger, and achieve the purpose of reducing the energy consumption of the data center.

Underwater Sensor Multi-Parameter Scheduling for Heterogenous Computing Nodes

Sensor-aware distributed workflow applications are becoming increasingly popular underwater. The apps are marine operations that generate data and process it based on its characteristics. Mobile-fog-cloud paradigms, as well as computing such as sensor nodes, have emerged. As previously stated, the nodes can be combined into a single system to achieve several goals. Many factors are considered, including network contents, workload fluctuation, variable execution durations, deadlines, and bandwidth. As a result, scheduling mobile workflow systems with multiple parameters might be challenging. The study suggests a novel content-efficient decision-aware task scheduling (CATSA) method for defining and adapting to complicated environmental changes. The CATSA consists of several components that work together to perform various benchmarks in the system, including a decision planner, sequencing, and scheduling. As evidenced by test findings during evaluation, the suggested architecture outperforms current studies regarding workflow execution quality of services and improved the makespan 30% and deadline meeting 40% in the study.

Obstetric Anesthesia Procedure-Based Workload and Facility Utilization of Society of Obstetric Anesthesia and Perinatology Centers of Excellence Designated Institutions.

Optimal workload and staffing for obstetric anesthesia services have yet to be determined. We surveyed Society of Obstetric Anesthesia and Perinatology (SOAP) Centers of Excellence (COE) for Obstetric Anesthesia Care institutions to evaluate procedure-based obstetric anesthesia workload and facility use. After institutional review board (IRB) exemption, an online survey instrument (REDCap) was sent by email (1 initial and 2 reminders) to all SOAP COEs. Survey data included the number of deliveries, cesarean delivery rate, neuraxial labor analgesia rate, the number of labor and operating rooms, and the number of in-house and backup obstetric anesthesia providers. Obstetric anesthesia activity was estimated using a time-based workload ratio per provider (Stanford Work Index, 1.0 = clinically working every minute of every hour on duty) during weekday, weeknight, and weekend shifts. We compared workload between academic and nonacademic centers and correlated operating and labor rooms with cesarean and vaginal delivery volume. Fifty-one of 53 surveys were returned (96% response rate). Data from 33 academic and 14 nonacademic US institutions were analyzed. For academic centers, median Stanford Work Index for all staff (included trainees and Certified Registered Nurse Anesthetists) was 0.20 (weekday) and 0.19 (weeknight and weekends); nonacademic centers were 0.33 (weekday, P < .001 versus academic), 0.23 (weeknight, P = .009 versus academic), and 0.23 (weekends, P = .03 versus academic practices). Attending-only Stanford Work Indices were similar between academic and nonacademic centers. Total number of rooms on the obstetric suite (operating, labor, or triage room) was strongly correlated with delivery volume ( R2 = 0.55). The results outline staffing procedure-based workload ratios and facility utilization at SOAP COEs in the United States. These data can be used by other institutions that provide obstetric anesthesia services to guide their obstetric anesthesia staffing. The importance of considering the workload associated with different shifts and between academic and nonacademic centers is also highlighted. The results show that approximately one-third of an obstetric anesthesiologist's workload is spent on performing procedures. We did not, however, measure the other tasks anesthesiologists practice as peripartum physicians (eg, managing critically ill parturients, doing pre- and postprocedural evaluations, or performing emergent and unexpected procedures), and future studies are required to determine the time required to perform these tasks. Studies to determine the optimal staffing models to handle workload fluctuations and improve outcomes are also required.

Elastic Resource Provisioning Using Data Clustering in Cloud Service Platform

Currently, cloud computing has received great attention in commerce and scientific research due to its flexibility and strong data processing capability. However, in view of the fact that the types of tasks display an upward trend as the growth of service demands, and the different types of tasks arrive at the system without regularity. Moreover, the resources deployed in cloud are insufficiency to be flexibly provisioned in the face of obvious workload fluctuations. In this article, we present a method of elastic resource provisioning using date clustering in cloud service platform. The framework of proposed method consists of three core components: tasks clustering, the amount of tasks prediction in cluster, dynamic resource provisioning and scheduling. In workload classification, we propose a clustering ensemble method, which utilizes a novel distance decision-making method to obtain the final results. Our method can effectively partition the arriving tasks into several clusters based on similarity among tasks. For each cluster, we forecast the amount of tasks arriving at next moment by prediction model based on time-series to provide reference for the follow-up resource provisioning. Afterwards, an energy-saving resource provisioning method is designed to dynamically provide resources for tasks in each cluster to meet their performance requirements. We implement the experiments in Google cloud traces dataset and the results show that our method achieves 92.3, 91.2 percent, and 3679.2 kW<inline-formula><tex-math notation="LaTeX">$ \cdot$</tex-math><alternatives><mml:math><mml:mo>&#x00B7;</mml:mo></mml:math><inline-graphic xlink:href="zhu-ieq1-3002755.gif"/></alternatives></inline-formula>h respectively in terms of guarantee ratio, resource utilization and total energy consumption, which demonstrates the effectiveness of proposed method for dynamic resource provisioning.

Automatic Evaluation and Comparison of Pub/Sub Systems Performance Improvements

Event-driven architectures are becoming more prevalent recently in multiple technological paradigms, especially in web applications, with message brokers being the cornerstone of these architectures. One of the most relevant implementations of these message brokers are content-based publish/subscribe systems. The performance of these systems is a critical factor for web engineering, since the web applications they support need to be reactive despite increases and fluctuations in workloads. However, an obstacle to the research of these systems is the lack of real and publicly available workloads, due to the privacy issue involved in disclosing the interests (subscriptions) of users and other commercial interests of the companies. In this paper we present a parameterizable automated system designed to syntactically translate workloads from different content-based publish/subscribe systems as a means to increase the availability of public workloads to solve the aforementioned problem. As a case study, we describe the evolution of a context-aware content-based publish/subscribe system (i.e. E-SilboPS) designed by the authors, which improves up to 5 times the performance of its previous version by reaching the maximum throughput limited by the physical resources of the hardware where it is deployed, as demonstrated by the conducted quantitative evaluation. Then, we validate the utility of the proposed automated workload generation system by using it to make the performance comparison between this new version E-SilboPS and one of the most cited publish/subscribe systems called PADRES, through a real trace of a massively multiplayer online game (MMOG) generated by the latter.

Exploiting Heterogeneity for Opportunistic Resource Scaling in Cloud-Hosted Applications

Cloud consumers have access to an increasingly diverse range of resource and contract options, but lack appropriate resource scaling solutions that can exploit this to minimize the cost of their cloud-hosted applications. Traditional approaches tend to use homogeneous resources and horizontal scaling to handle workload fluctuations and do not leverage resource and contract heterogeneity to optimize cloud costs. In this paper, we propose a novel opportunistic resource scaling approach that exploits both resource and contract heterogeneity to achieve cost-effective resource allocations. We model resource allocation as an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">unbounded knapsack problem</i> , and resource scaling as an <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">one-step ahead resource allocation problem</i> . Based on these models, we propose two scaling strategies: (a) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">delta capacity optimization</i> , which focuses on optimizing costs for the difference between existing resource allocation and the required capacity based on the forecast workload, and (b) <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">full capacity optimization</i> , which focuses on optimizing costs for resource capacity corresponding to the forecast workload. We evaluate both strategies using two real world workload datasets, and compare them against three different scaling strategies. The results show that our proposed approach, particularly full capacity optimization, outperforms all of them and offers in excess of 70 percent cost savings compared to the traditional scaling approach.

A Port Loading Model to Minimize Workload Fluctuation for Surface Ship Maintenance Jobs

A Port Loading Model to Minimize Workload Fluctuation for Surface Ship Maintenance Jobs

A novel integrated fuzzy control system toward automated local airflow management in data centers

Today, Data Centers (DCs) are dynamic environments with considerable fluctuations in workload and power dissipation. As a result, active monitoring and dynamic thermal management strategies are essential. In this study, an automated dynamic airflow management technique using air dampers was introduced to manage cold air delivery to individual aisles based on the Information Technology Equipment (ITE) airflow demand. Within this management system, pressure measurements inside the Cold Aisle Containment (CAC) and the plenum were considered controlled variables. First, a feedback fuzzy controller was designed to regulate the airflow delivered to the aisles by adjusting the Open Area Ratio (OAR) of the air dampers. Then, to improve the system’s performance and to implement a control system which was adaptable to environmental changes, another fuzzy controller was developed to adjust the blower speed of the cooling units. To estimate the required airflow for provisioning all the ITE in a DC, an Artificial Neural Network (ANN) was developed to characterize the air dampers. This study experimentally examined several opportunities for improving the thermal management and energy performance of DCs with automatic control schemes. Experimental data showed that by using the proposed cooling control strategy, 75% of the cooling units’ blower powers and 16% of the chiller’s power were saved while maintaining proper thermal management conditions compared to the worst case scenario in which the air dampers were completely open and the cooling units’ blower speeds were at maximum. Experimental data from the implementation of the holistic control methodology indicated that there was minimal air leakage from the plenum to the room. Additionally, this approach achieved more efficient airflow delivery from CRAH units to the cold aisles with minimal air loss through leakage.

Real-time prediction of short-timescale fluctuations in cognitive workload

Human operators often experience large fluctuations in cognitive workload over seconds timescales that can lead to sub-optimal performance, ranging from overload to neglect. Adaptive automation could potentially address this issue, but to do so it needs to be aware of real-time changes in operators’ spare cognitive capacity, so it can provide help in times of peak demand and take advantage of troughs to elicit operator engagement. However, it is unclear whether rapid changes in task demands are reflected in similarly rapid fluctuations in spare capacity, and if so what aspects of responses to those demands are predictive of the current level of spare capacity. We used the ISO standard detection response task (DRT) to measure cognitive workload approximately every 4 s in a demanding task requiring monitoring and refueling of a fleet of simulated unmanned aerial vehicles (UAVs). We showed that the DRT provided a valid measure that can detect differences in workload due to changes in the number of UAVs. We used cross-validation to assess whether measures related to task performance immediately preceding the DRT could predict detection performance as a proxy for cognitive workload. Although the simple occurrence of task events had weak predictive ability, composite measures that tapped operators’ situational awareness with respect to fuel levels were much more effective. We conclude that cognitive workload does vary rapidly as a function of recent task events, and that real-time predictive models of operators’ cognitive workload provide a potential avenue for automation to adapt without an ongoing need for intrusive workload measurements.

An Elasticity Framework for Distributed Message Queuing Telemetry Transport Brokers

An Elasticity Framework for Distributed Message Queuing Telemetry Transport Brokers

A scalable and low latency probe-based scheduler for data analytics frameworks

Today’s data analytics frameworks divide jobs into many parallel tasks such that each task operates on a small partition of data in order to execute jobs with low latency. Such frameworks often rely on probe-based distributed schedulers to tackle the challenge of reducing the associated overhead. Unfortunately, the existing solutions do not perform efficiently under workload fluctuations and heterogeneous job durations. This is due to a problem called Head-of-Line blocking, i.e., short tasks are enqueued at workers behind longer tasks. To overcome this problem, we propose Peacock (Khelghatdoust and Gramoli, 0000) [25] a new fully distributed probe-based scheduling method. Unlike the existing methods, Peacock introduces a novel probe rotation technique. Workers form a ring overlay network and rotate probes using elastic queues of workers. It is augmented by a novel starvation-free probe reordering algorithm executed by workers. We evaluate Peacock against two existing state-of-the-art probe based solutions through a trace driven simulation of up to 20,000 workers and a distributed experiment of 100 workers in Apache Spark under Google, Cloudera, and Yahoo! traces. The performance results indicate that Peacock outperforms the state-of-the-art in all cluster sizes and loads. Our distributed experiments confirm our simulation results.

Low overhead performance monitoring for shared infrastructures

Modern systems make use of virtualization to build scalable and elastic applications. A major example is cloud computing, which has emerged in the last decades as a cost-effective paradigm for hosting and delivering services over the Internet. Cloud providers adopt server consolidation strategies to manage resources, reducing underutilized hosts. However, collocated applications compete for physical resources. So, their particularities, as well as workload fluctuations, may negatively affect cloud applications’ performance. Performance monitors are useful tools for tracking performance behavior. The observation of events occurring in the system provides a detailed view of how applications affect resource usage. This way, performance counters can assist developers and enable the detection of bottlenecks, allowing administrators to verify whether service level agreements are met. However, the monitoring itself can be a contention source, especially in large scale systems, where hundreds of performance counters are collected, and the acquired data are stored at a high frequency. In this work, the influence of monitoring overhead in virtualized environments is analyzed. As a mean to reduce shared resource contention, it is proposed an approach to alleviate monitoring costs, bringing as benefits: an automatic method for performance counters selection; and overhead reduction by discarding unnecessary performance counters. Through experimental evaluation, this work characterizes the monitoring overhead in virtualized environments. A performance comparison using both the complete and the reduced set of performance counters highlights the benefits of adopting the proposed approach.

Analysis of Dynamic Changes in Cognitive Workload During Cardiac Surgery Perfusionists' Interactions With the Cardiopulmonary Bypass Pump.

This novel preliminary study sought to capture dynamic changes in heart rate variability (HRV) as a proxy for cognitive workload among perfusionists while operating the cardiopulmonary bypass (CPB) pump during real-life cardiac surgery. Estimations of operators' cognitive workload states in naturalistic settings have been derived using noninvasive psychophysiological measures. Effective CPB pump operation by perfusionists is critical in maintaining the patient's homeostasis during open-heart surgery. Investigation into dynamic cognitive workload fluctuations, and their relationship with performance, is lacking in the literature. HRV and self-reported cognitive workload were collected from three Board-certified cardiac perfusionists (N = 23 cases). Five HRV components were analyzed in consecutive nonoverlapping 1-min windows from skin incision through sternal closure. Cases were annotated according to predetermined phases: prebypass, three phases during bypass, and postbypass. Values from all 1min time windows within each phase were averaged. Cognitive workload was at its highest during the time between initiating bypass and clamping the aorta (preclamp phase during bypass), and decreased over the course of the bypass period. We identified dynamic, temporal fluctuations in HRV among perfusionists during cardiac surgery corresponding to subjective reports of cognitive workload. Not only does cognitive workload differ for perfusionists during bypass compared with pre- and postbypass phases, but differences in HRV were also detected within the three bypass phases. These preliminary findings suggest the preclamp phase of CPB pump interaction corresponds to higher cognitive workload, which may point to an area warranting further exploration using passive measurement.

Automated Fine-Grained CPU Cap Control in Serverless Computing Platform

Serverless computing has emerged as a new cloud computing execution model that liberates users and application developers from explicitly managing `physical' resources, leaving such a resource management burden to service providers. In this article, we study the problem of resource allocation for multi-tenant serverless computing platforms explicitly taking into account workload fluctuations including sudden surges. In particular, we investigate different root causes of performance degradation in these platforms where tenants (their applications) have different workload characteristics. To this end, we develop a fine-grained CPU cap control solution as a resource manager that dynamically adjusts CPU usage limit (or CPU cap) concerning applications with same/similar performance requirements, i.e., application groups. The adjustment of CPU caps applies primarily to co-located worker processes of serverless computing platforms to minimize resource contention, which is the major source of performance degradation. The actual adjustment decisions are made based on performance metrics (e.g., throttled time and queue length) using a group-aware scheduling algorithm. The extensive experimental results performed in our local cluster confirm that the proposed resource manager can effectively eliminate the burden of explicit reservation of computing capacity, even when fluctuations and sudden surges in the incoming workload exist. We measure the robustness of the proposed resource manager by comparing it with several heuristics which extensively used in practice, including the enhanced version of round robin and the least length queue scheduling policies, under various workload intensities driven by real-world scenarios. Notably, our resource manager outperforms other heuristics by decreasing skewness and average response time up to 44 and 94 percent, respectively, while it does not over-use the CPU resources.

Resource provisioning for IoT services in the fog computing environment: An autonomic approach

In the recent years, the Internet of Things (IoT) services has been increasingly applied to promote the quality of the human life and this trend is predicted to stretch for into future. With the recent advancements in IoT technology, fog computing is emerging as a distributed computing model to support IoT functionality. Since the IoT services will experience workload fluctuations over time, it is important to automatically provide the proper number of sufficient fog resources to address the workload changes of IoT services to avoid the over- or under-provisioning problems, meeting the QoS requirements at the same time. In this paper, an efficient resource provisioning approach is presented. This approach is inspired by autonomic computing model using Bayesian learning technique to make decisions about the increase and decrease in the dynamic scaling fog resources to accommodate the workload from IoT services in the fog computing environment. Also, we design an autonomous resource provisioning framework based on the generic fog environment three-tier architecture. Finally, we validate the effectiveness of our solution under three workload traces. The simulation results indicate that the proposed solution reduces the total cost and delay violation, and increases the fog node utilization compared with the other methods.

An Artificial Neural Network Approach to Power Consumption Model Construction for Servers in Cloud Data Centers

The power consumption estimation or prediction of cloud servers is the basis of energy-aware scheduling to realize energy saving in cloud datacenters. The existing works are mainly based on the static mathematical formulas which establish the relationship between the server power consumption and the system performance. However, these models are weak in adaptability and generalization ability, not adaptable to the changes and fluctuation of different workload, and demanding on the clear and profound understanding of the inner relationship among related power consumption parameters. Therefore, we propose the ANN (Artificial Neural Network) method to model the power consumption of the servers in datacenters, a kind of end-to-end black box model. We performed a fine-grained and in-depth analysis about the system performance and power consumption characteristics of the CPU, memory, and disk of the server running different types of task loads, and selected a set of performance counters that can fully reflect the status of system power consumption as the input of the model. Then, we establish power consumption models based on BP neural network, Elman neural network, and LSTM neural network, respectively. In order to get a better result, we use data collected from four different types of task loads (i.e., CPU-intensive, memory-intensive, I/O-intensive, and mixed load) to train, validate, and test our target models. The experimental results show that, compared with multiple linear regression and support vector regression, the proposed three power models have better performance in predicting the server's real-time power consumption.

Cost‐aware cloud bursting in a fog‐cloud environment with real‐time workflow applications

SummaryCloud bursting is a concept originating from the hybrid cloud computing paradigm. During workload spikes, the local resources of the private cloud are supplemented by resources in the public cloud. This technique could also be applied in a fog computing environment, in order to handle workload fluctuations, by offloading applications to the cloud. Toward this direction, in this article, we propose a strategy for the utilization of supplementary cloud resources, in order to assist in the processing of Internet of Things workflow jobs that arrive dynamically in a fog environment. As the cloud involves higher data transfer latency and monetary cost, our approach takes into account these two factors, in addition to the real‐time constraints of the workload. The proposed scheduling heuristic is based on the tradeoff between performance and monetary cost. During resource selection, different contribution factors of these two parameters are assessed. Furthermore, the proposed scheduling method is compared against a baseline policy that utilizes only the fog resources. The simulation experiments were carried out under different sizes of workflow input data and for workloads with soft and hard deadlines.

Bullfighting extreme scenarios in efficient hyper-scale cluster computing

Data centres are quickly evolving to support new demands for Cloud-Computing services. Extreme workload peaks represent a challenge for the maintenance of the performance and service level agreements, even more when operation costs need to be minimised. In this paper, we first present an extensive analysis of the impact of extreme workloads in large-scale realistic Cloud-Computing data centres, including a comparison between the most relevant centralised resource-managing models. Moreover, we extend our previous works by proposing a new energy-efficiency policy called Bullfighter which is able to keep performance key performance indicators while reducing energy consumption in extreme scenarios. This policy employs queue-theory distributions to foresee workload demands and adapt automatically to workload fluctuations even in extreme environments, while avoiding the fine-tuning required for other energy policies. Finally, it is shown through extensive simulation that Bullfighter can save more than 40% of energy in the aforementioned scenarios without exerting any noticeable impact on data-centre performance.

Designing Novel Therapies to Mend Broken Hearts: ATF6 and Cardiac Proteostasis.

The heart exhibits incredible plasticity in response to both environmental and genetic alterations that affect workload. Over the course of development, or in response to physiological or pathological stimuli, the heart responds to fluctuations in workload by hypertrophic growth primarily by individual cardiac myocytes growing in size. Cardiac hypertrophy is associated with an increase in protein synthesis, which must coordinate with protein folding and degradation to allow for homeostatic growth without affecting the functional integrity of cardiac myocytes (i.e., proteostasis). This increase in the protein folding demand in the growing cardiac myocyte activates the transcription factor, ATF6 (activating transcription factor 6α, an inducer of genes that restore proteostasis. Previously, ATF6 has been shown to induce ER-targeted proteins functioning primarily to enhance ER protein folding and degradation. More recent studies, however, have illuminated adaptive roles for ATF6 functioning outside of the ER by inducing non-canonical targets in a stimulus-specific manner. This unique ability of ATF6 to act as an initial adaptive responder has bolstered an enthusiasm for identifying small molecule activators of ATF6 and similar proteostasis-based therapeutics.

Dynamic Provisioning of Service Composition in a Multi-Tenant SaaS Environment

Multi-tenant service composition has become a common delivery model for business processes in cloud computing. To dynamically support the workload tenant variation, elasticity holds the promise of ensuring the quality of service (QoS) of the business process by providing the involved service instances at a low cost. However, integrating both of multi-tenancy and elasticity during service composition is a key problem for serving multiple tenants from a single process instance. Nowadays, existing approaches in the field of cloud service composition, although numerous, still fall short since they cannot adequately address issues related to supporting the scalability of the composed service and adapting it to the workload fluctuation. In this paper, we propose a holistic approach which makes the dynamic multi-tenant services matching and manages their elasticity in distributed business processes. This approach is based on a generic service pattern that integrates multi-tenancy property and handles elasticity at the process and service levels. Furthermore, we present elastic composition algorithms to compose multi-tenant cloud services and perform their elasticity through the proposed service pattern. The evaluation of our approach, compared to the baseline approach, proves that the latency taken to provide an elastic multi-tenant service composition and detect its SLA (Service Level Agreements) violation are reasonably short. We also show that the CPU overhead of using our approach is negligible. Furthermore, experimental results demonstrate the merits of our approach in terms of minimizing the memory consumption through the deployed service instances.