Self‐adaptive workload classification and forecasting for proactive resource provisioning

Cloud platforms are increasingly using containers for lightweight virtualization. However, the mainstream operating systems are currently limited in their capabilities in customizing containers’ resource management. There remains two main challenges in resource allocations. First, the application workloads can be time-varying, leading to a problem of resource over- or under-allocations. Second, it becomes difficult to minimize the resource provisioning cost while guaranteeing the SLO (service level objective). To address these challenges, we propose ConLAR, a learning-to-allocate approach that predicts and allocates resources to Docker containers under time-varying workloads. ConLAR efficiently reduces over-provisioning cost with the SLO guarantees by taking an Observing-Predicting-Allocating-Executing paradigm: given a container, it observes the running of the online application and its environment, leverages the LSTM (long short term memory) model to predict its future workload, adaptively learns to construct resource allocation strategies with two objectives through RL (reinforcement learning), and executes them to scale container resources dynamically. We have evaluated ConLAR on two real-world workloads of ClarkNet and GoogleClusterData. The results clearly show the effectiveness of ConLAR. In particular, ConLAR achieves a resource over-provisioning cost of less than 16.5% and an SLO violations rate of 8.9%; it also shows good flexibility to learn different adaption policies.

Reliable long-term streamflow forecasting is invaluable for water resource planning and management which allocates water supply according to the demand of water users. It is necessary to get probabilistic forecasts to establish risk-based reservoir operation policies. Probabilistic forecasts may be useful for the users who assess and manage risks according to decision-making responding forecasting results. Probabilistic forecasting of seasonal inflow to Andong dam is performed and assessed using selected predictors from sea surface temperature and 500 hPa geopotential height data. Categorical probability forecast by Piechota's method and logistic regression analysis, and probability forecast by conditional probability density function are used to forecast seasonal inflow. Kernel density function is used in categorical probability forecast by Piechota's method and probability forecast by conditional probability density function. The results of categorical probability forecasts are assessed by Brier skill score. The assessment reveals that the categorical probability forecasts are better than the reference forecasts. The results of forecasts using conditional probability density function are assessed by qualitative approach and transformed categorical probability forecasts. The assessment of the forecasts which are transformed to categorical probability forecasts shows that the results of the forecasts by conditional probability density function are much better than those of the forecasts by Piechota's method and logistic regression analysis except for winter season data.

Abstract: Online resource management in cloud computing must respond to rapidly changing demand and system conditions, often exceeding the adaptability of static or periodically recomputed optimisation plans. This study formulates cloud resource management as a Markov decision process (MDP) and investigates deep reinforcement learning (DRL) as an approximate dynamic programming (ADP) approach for deriving adaptive control policies. The system state encompasses workload, capacity utilisation, and performance indicators, while actions govern virtual machine placement, scaling, and server power-state transitions. Reward functions combine energy and operating costs with penalties associated with service-level agreement violations. DRL algorithms, including deep Q-networks, advantage actor–critic, and proximal policy optimization, are implemented in PyTorch and trained through interaction with a CloudSim-based simulation environment calibrated to realistic load and power profiles. The learned policies are evaluated against deterministic and scenario-based stochastic optimisation baselines using metrics such as average cost, energy consumption, SLA violations, and robustness to non-stationary demand. Results show that DRL policies can approach or surpass the performance of stochastic optimisation solutions in highly dynamic scenarios, while maintaining feasible control actions. The study demonstrates how DRL can be positioned within an OR/ADP framework and highlights conditions under which learning-based methods offer significant advantages for online cloud control. Keywords: deep reinforcement learning, approximate dynamic programming, cloud computing, Markov decision processes, online resource management, energy-aware control, PyTorch, simulation

We use statistical learning to estimate end-to-end QoS metrics from device statistics, collected from a server cluster and an OpenFlow network. The results from our testbed, which runs a video-on-demand service and a key-value store, demonstrate that the learned models can estimate QoS metrics like frame rate or response time with errors bellow 10% for a given client. Interestingly, we find that service-level QoS metrics seem encoded in network statistics and it suffices to collect OpenFlow per port statistics to achieve accurate estimation at small overhead for data collection and model computation.

Service providers want to reduce datacenter costs by consolidating workloads onto fewer servers. At the same time, customers have performance goals, such as meeting tail latency Service Level Objectives (SLOs). Consolidating workloads while meeting tail latency goals is challenging, especially since workloads in production environments are often bursty. To limit the congestion when consolidating workloads, customers and service providers often agree upon rate limits. Ideally, rate limits are chosen to maximize the number of workloads that can be co-located while meeting each workload's SLO. In reality, neither the service provider nor customer knows how to choose rate limits. Customers end up selecting rate limits on their own in some ad hoc fashion, and service providers are left to optimize given the chosen rate limits. This paper describes WorkloadCompactor, a new system that uses workload traces to automatically choose rate limits simultaneously with selecting onto which server to place workloads. Our system meets customer tail latency SLOs while minimizing datacenter resource costs. Our experiments show that by optimizing the choice of rate limits, WorkloadCompactor reduces the number of required servers by 30--60% as compared to state-of-the-art approaches.

Resource under-utilization is prevalent in data centers as a result of substantial stochastic workload variations and resource provision guarantee for the requirements to provide high-level service availability. To avoid resource under-utilization, recently, the Economy cloud computing service class (Economy class in short) is proposed that sells allocated but unused (i.e., reclaimed) physical resources with long-term service level objectives (SLOs). Though the Economy class would lead to revenue benefit, how much of the reclaimed resources should be used as the Economy class (i.e., capacity planning) and how to set its price and SLAs to maximize revenue benefit have not been studied. To address this problem, in this paper, we propose a multi-level SLA for Economy class, which has multiple levels of SLOs and penalty. We then present a rigorous theoretical analysis on the capacity planning and pricing of the new Economy class with an objective to maximize the revenue of the cloud computing service provider. Theoretical results are validated using resource utilization data derived from a publicly available Google cluster workload trace dataset.

Virtualization provides increased control and flexibility on how resources are allocated to applications. However, common resource provisioning mechanisms do not fully use these advantages; either they provide limited support for applications demanding quality of service, or the resource allocation complexity is high. To address these issues we developed Themis, a market-based application management platform. By limiting the coupling between the applications and resource management, Themis can support diverse types of applications and performance goals while ensuring maximized resource usage. In this paper we present the performance of Themis when users execute batch applications with different Service Level Objectives such as deadlines.

Capacity planning for IaaS cloud providers offering multiple service classes

Cloud service is being adopted as a utility for large numbers of tenants by renting Virtual Machines (VMs). But for cloud storage, unpredictable IO characteristics make accurate Service-Level-Objective (SLO) enforcement challenging. As a result, it has been very difficult to support simple-to-use and technology-agnostic SLO specifying a particular value for a specific metric (e.g., storage bandwidth). This is because the quality of SLO enforcement depends on performance error and fluctuation that measure the precision of SLO enforcement . High precision of SLO enforcement is critical for user-oriented performance customization and user experiences. To address this challenge, this article presents V-Cup, a framework for VM-oriented customizable SLO and its near-precise enforcement. It consists of multiple auto-tuners, each of which exports an interface for a tenant to customize the desired storage bandwidth for a VM and enable the storage bandwidth of the VM to converge on the target value with a predictable precision. We design and implement V-Cup in the Xen hypervisor based on the fair sharing scheduler for VM-level resource management. Our V-Cup prototype evaluation shows that it achieves satisfying performance guarantees through near-precise SLO enforcement.

Consolidating applications of conflicting service level objectives (SLOs) to share virtualized resources in cloud datacenters requires efficient resource management to ensure overall high Quality-of-Service (QoS). Applications of different performance targets often exhibit different resource demands. Thus, it is not trivial to translate individual application SLOs to corresponding resource shares in a shared virtualized environment to meet performance targets. In this paper, we present CtrlCloud, a performance-aware resource controlling system, that adaptively allocates resources, with a resource-share controller and an allocation optimization model. The controller automatically adapts resource demands based on performance deviations, while the optimization model resolves conflicts in resource demands from multiple co-located applications based on their ongoing performance achieved. We implement a proof-of-concept prototype of CtrlCloud in Python on top of Xen hypervisor. Our experimental results indicate that CtrlCloud can optimize allocations of CPU resources across multiple applications to maintain the 95th percentile latency within predefined SLO targets. CtrlCloud also provides QoS differentiation and yet fulfilling of CPU share demands from applications is maximized given resource availability. We further compare CtrlCloud against two other resource allocation methods commonly used in current clouds. CtrlCloud improves resource utilization by allocating resource shares optimal to 'actual needs' as it employs share-performance online modeling.

Occupancy rate is a critical factor in a hotel because it is used to measure the operational success of a hotel. The higher the hotel occupancy rate, the more successful the hotel business is in generating revenue. In the hospitality business, occupancy rates are very important to monitor and analyse as marketing strategies and pricing policies. This study compares the forecasting of occupancy rates at Hotel X using neural network method and decision tree method. The dataset used in this study is room available, room sold, and available occupancy percentage data at Hotel X from April 2018 to June 2023. The simulation was carried out by dividing the data into training data and testing data with a percentage ratio of 70:30, 75:25, 80:20, 85:15, and 90:10. The forecasting results that have been carried out using a neural network with one hidden layer have an optimal RSME result of 0.010 for split data of 70%:30% and 80%:20% while using a neural network with two hidden layers the optimal result of RSME is 0.013 for split data of 75%:25%. Forecasting results using decision tree RSME optimal results of 0.022 for split data 85%:15% and 90%:10%. From this forecasting, the most optimal results use a neural network with one hidden layer for data splits of 70%:30% and 80%:20% with RSME results of 0.010. The results of the research can be used by Hotel X as a policy determination in the next hotel management.

Supercomputers play a pivotal role in advancing research and development across diverse scientific and engineering domains. However, configuring resource management in these systems to ensure maximum productivity and cost-effectiveness is a challenge. Workload simulation emerges as a crucial tool in this context, offering a mechanism to explore resource management configurations in the presence of expected user behaviors. This paper focuses on a specific requirement for simulation-based optimization applied to tuning resource management configurations: the need for simulators that are both precise and efficient. This paper introduces a job shaping strategy to accommodate real workload traces under varying resource management policies in discrete-event RMS simulations. Our findings from evaluating the proposed strategy on a real-world case study suggest that job shaping allows effectively capturing changes in system behavior, regardless of whether some of the real workload traces used as input to the simulation are incompatible with the simulated policies.

The Egyptian cotton crop is considered one of the important strategic crops, as it is one of the main pillars of the Egyptian national economic structure, and is used in many local industries, such as the textile industry, oil and soap industry, animal feed etc. Also, the cultivation of this crop employs more than one million workers and earns the Country a lot of foreign exchange from the export of this crop. In the past period, there has been a steady decline in the cultivation; in terms of land area and quantity produced of the crop. To achieve the Aims of this study Used ARIMA models because these models combine the method of Autoregression, the moving average of the time series to the forecasting of the cultivated area, Productivity, and Production of the cotton crop where these models are characterized by high accuracy in the analysis of Time series. We found from the results the cultivated area of that crop reached 230.30 thousand acres in 2018 then decreased to 27.85 thousand acres in 2024. After that the Production of the cotton crop in Egypt will cease because there is no area planted of cotton crop in Egypt after the year 2024. Also, the results of the productivity forecast were 7.06 quintals in 2018 then dropped to 6.61 quintals in 2025. Also, the results of the forecast of Production of the cotton crop reached 1173.61 thousand quintals in 2018 then reduced to 128.33 thousand quintals in 2024. Then after 2024, the Production of Egyptian cotton crop ceased permanently. Keywords: forecasting, ARIMA model, Moving average model, Cotton crop. Auto-regressive model. Egypt, Auto-correlation function, Partial Auto-correlation Function.

Service level agreements (SLAs) gain more and more importance in the area of cloud computing. An SLA is a contract between a customer and a cloud service provider (CSP) in which the CSP guarantees functional and non-functional quality of service parameters for cloud services. Since CSPs have to pay for the hardware used as well as penalties for violating SLAs, they are eager to fulfill these agreements while at the same time optimizing the utilization of their resources.In this paper we examine SLA-aware VM scheduling strategies for cloud data centers. The service level objectives considered are resource usage and availability. The sample resources are CPU and RAM. They can be overprovisioned by the CSPs which is the main leverage to increase their revenue. The availability of a VM is affected by migrating it within and between data centers. To get realistic results, we simulate the effect of the strategies using the FederatedCloudSim framework and real-world workload traces of business-critical VMs.Our evaluation shows that there are considerable differences between the scheduling strategies in terms of SLA violations and the number of migrations. From all strategies considered, the combination of the Minimization of Migrations strategy for VM selection and the Worst Fit strategy for host selection achieves the best results.

In the context of digital development, society is rapidly advancing in digital technologies, resulting in the accumulation of large amounts of data on consumer behaviour. At the same time, effective analysis of information is becoming a critical factor for business success and informed management decisions. The aim of the work is to conduct a comprehensive review and systematisation of existing data analysis methods for determining demand for goods, to identify the optimal set of methods for developing a system for analysing and forecasting demand for computer equipment. The research methodology is based on a systematic approach to the analysis of scientific sources and a comparative analysis of data processing methods. The work applies methods of theoretical generalisation to systematise approaches to data analysis, including descriptive statistics (mean, median, mode, variance, standard deviation, asymmetry and excess coefficients), correlation analysis (Pearson and Spearman coefficients), machine learning methods (k-means and hierarchical clustering, dimension reduction methods, the Apriori method, the support vector method, decision trees) and time series analysis. A structural analysis of the advantages and limitations of each method in the context of forecasting demand for goods was conducted. The scientific novelty lies in the development of a comprehensive methodological approach to analysing demand for computer equipment, which integrates time series analysis to identify trends, seasonal analysis to identify cyclical patterns, descriptive statistics to assess demand variability, and interactive visualisation for effective interpretation of results. Conclusions. It has been established that for effective analysis and forecasting of demand for computer equipment, it is necessary to comprehensively apply methods of time series analysis, descriptive statistics, dimension reduction methods, seasonal analysis, and interactive visualisation.