Optimizing Resource Allocation Framework for Multi-Cloud Environment

Cloud computingmakes dynamic resource provisioning more accessible.Monitoring a functioning service is crucial,and changes are made when particular criteria are surpassed.This research explores the decentralized multi-cloud environment for allocating resources and ensuring the Quality of Service(QoS),estimating the required resources,and modifying allotted resources depending on workload and parallelism due to resources.Resource allocation is a complex challenge due to the versatile service providers and resource providers.The engagement of different service and resource providers needs a cooperation strategy for a sustainable quality of service.The objective of a coherent and rational resource allocation is to attain the quality of service.It also includes identifying critical parameters to develop a resource allocation mechanism.A framework is proposed based on the specified parameters to formulate a resource allocation process in a decentralized multi-cloud environment.The three main parameters of the proposed framework are data accessibility,optimization,and collaboration.Using an optimization technique,these three segments are further divided into subsets for resource allocation and long-term service quality.The CloudSim simulator has been used to validate the suggested framework.Several experiments have been conducted to find the best configurations suited for enhancing collaboration and resource allocation to achieve sustained QoS.The results support the suggested structure for a decentralized multi-cloud environment and the parameters that have been determined.

Cloud Resource Integrated Prediction Model Based on Variational Modal Decomposition-Permutation Entropy and LSTM

Predicting the usage of container cloud resources has always been an important and challenging problem in improving the performance of cloud resource clusters.We proposed an integrated prediction method of stacking container cloud resources based on variational modal decomposition(VMD)-Permutation entropy(PE)and long short-term memory(LSTM)neural network to solve the prediction difficulties caused by the non-stationarity and volatility of resource data.The variational modal decomposition algorithm decomposes the time series data of cloud resources to obtain intrinsic mode function and residual components,which solves the signal decomposition algorithm’s end-effect and modal confusion problems.The permutation entropy is used to evaluate the complexity of the intrinsic mode function,and the reconstruction based on similar entropy and low complexity is used to reduce the difficulty of modeling.Finally,we use the LSTM and stacking fusion models to predict and superimpose;the stacking integration model integrates Gradient boosting regression(GBR),Kernel ridge regression(KRR),and Elastic net regression(ENet)as primary learners,and the secondary learner adopts the kernel ridge regression method with solid generalization ability.The Amazon public data set experiment shows that compared with Holt-winters,LSTM,and Neuralprophet models,we can see that the optimization range of multiple evaluation indicators is 0.338∼1.913,0.057∼0.940,0.000∼0.017 and 1.038∼8.481 in root means square error(RMSE),mean absolute error(MAE),mean absolute percentage error(MAPE)and variance(VAR),showing its stability and better prediction accuracy.

Formal Modeling and Discovery of Multi-instance Business Processes: A Cloud Resource Management Case Study

Process discovery, as one of the most challenging process analysis techniques, aims to uncover business process models from event logs. Many process discovery approaches were invented in the past twenty years;however, most of them have difficulties in handling multi-instance sub-processes. To address this challenge, we first introduce a multi-instance business process model(MBPM) to support the modeling of processes with multiple sub-process instantiations. Formal semantics of MBPMs are precisely defined by using multi-instance Petri nets(MPNs)that are an extension of Petri nets with distinguishable tokens.Then, a novel process discovery technique is developed to support the discovery of MBPMs from event logs with sub-process multi-instantiation information. In addition, we propose to measure the quality of the discovered MBPMs against the input event logs by transforming an MBPM to a classical Petri net such that existing quality metrics, e.g., fitness and precision, can be used.The proposed discovery approach is properly implemented as plugins in the Pro M toolkit. Based on a cloud resource management case study, we compare our approach with the state-of-theart process discovery techniques. The results demonstrate that our approach outperforms existing approaches to discover process models with multi-instance sub-processes.

Resource pre-allocation algorithms for low-energy task scheduling of cloud computing

In order to lower the power consumption and improve the coefficient of resource utilization of current cloud computing systems, this paper proposes two resource pre-allocation algorithms based on the ＂shut down the redundant, turn on the demanded＂ strategy here. Firstly, a green cloud computing model is presented, abstracting the task scheduling problem to the virtual machine deployment issue with the virtualization technology. Secondly, the future workloads of system need to be predicted： a cubic exponential smoothing algorithm based on the conservative control（CESCC） strategy is proposed, combining with the current state and resource distribution of system, in order to calculate the demand of resources for the next period of task requests. Then, a multi-objective constrained optimization model of power consumption and a low-energy resource allocation algorithm based on probabilistic matching（RA-PM） are proposed. In order to reduce the power consumption further, the resource allocation algorithm based on the improved simulated annealing（RA-ISA） is designed with the improved simulated annealing algorithm. Experimental results show that the prediction and conservative control strategy make resource pre-allocation catch up with demands, and improve the efficiency of real-time response and the stability of the system. Both RA-PM and RA-ISA can activate fewer hosts, achieve better load balance among the set of high applicable hosts, maximize the utilization of resources, and greatly reduce the power consumption of cloud computing systems.

Resource Based Automatic Calibration System (RBACS) Using Kubernetes Framework

Kubernetes,a container orchestrator for cloud-deployed applications,allows the application provider to scale automatically to match thefluctuating intensity of processing demand.Container cluster technology is used to encapsulate,isolate,and deploy applications,addressing the issue of low system reliability due to interlocking failures.Cloud-based platforms usually entail users define application resource supplies for eco container virtualization.There is a constant problem of over-service in data centers for cloud service providers.Higher operating costs and incompetent resource utilization can occur in a waste of resources.Kubernetes revolutionized the orchestration of the container in the cloud-native age.It can adaptively manage resources and schedule containers,which provide real-time status of the cluster at runtime without the user’s contribution.Kubernetes clusters face unpredictable traffic,and the cluster performs manual expansion configuration by the controller.Due to operational delays,the system will become unstable,and the service will be unavailable.This work proposed an RBACS that vigorously amended the distribution of containers operating in the entire Kubernetes cluster.RBACS allocation pattern is analyzed with the Kubernetes VPA.To estimate the overall cost of RBACS,we use several scientific benchmarks comparing the accomplishment of container to remote node migration and on-site relocation.The experiments ran on the simulations to show the method’s effectiveness yielded high precision in the real-time deployment of resources in eco containers.Compared to the default baseline,Kubernetes results in much fewer dropped requests with only slightly more supplied resources.

Characteristics of Cloud Water Resource and Precipitation Efficiency of Hydrometeors over Northwest China

Understanding the characteristics of cloud water resource(CWR)and precipitation efficiency of hydrometeors(PEh)is imperative for the application of CWR in Northwest China.The atmospheric precipitable water(PW)in all four seasons and clouds and PEh in summer were studied with ERA-5 and CloudSat data in this region.The results show that topography,especially in the Tibetan Plateau,exerts significant impacts on the precipitation and PW in summer,since large amounts of clouds are distributed along the mountain ranges.The study region is divided into four typical areas:the monsoon area in eastern Northwest China(NWE),the Qilian Mountains area(QM),the Tianshan Mountains area(TM),and the Source of Three Rivers area(STR).Over the four areas,cloud top height(6.3 km)and cloud base height(3.3 km)over NWE are higher,and precipitating clouds are thicker(7 km)in the single-layer clouds.Liquid water content decreases with increasing altitude,while the ice water content first increases and then decreases.Liquid water path is higher over NWE(0.11 kg m^(−2))than over TM and STR(0.05 kg m^(−2)),and the ice water path is mainly concentrated within the range of 0.025–0.055 kg m^(−2).The PEh values are distributed unevenly and affected evidently by the terrain.Although the PEh values in the four typical areas(0.3–0.6)are higher than those in other regions,the CWR is relatively abundant and has a higher exploitation potential.Therefore,it is well-founded to exploit CWR for alleviating water shortages in these areas of Northwest China in summer.

Quantifying the Cloud Water Resource:Basic Concepts and Characteristics

The water in the air is composed of water vapor and hydrometeors,which are inseparable in the global atmosphere.Precipitation basically comes from hydrometeors instead of directly from water vapor,but hydrometeors are rarely focused on in previous studies.When assessing the maximum potential precipitation,it is necessary to quantify the total amount of hydrometeors present in the air within an area for a certain period of time.Those hydrometeors that have not participated in precipitation formation in the surface,suspending in the atmosphere to be exploited,are defined as the cloud water resource(CWR).Based on the water budget equations,we defined 16 terms(including 12 independent ones)respectively related to the hydrometeors,water vapor,and total water substance in the atmosphere,and 12 characteristic variables related to precipitation and CWR such as precipitation efficiency(PE)and renewal time(RT).Correspondingly,the CWR contributors are grouped into state terms,advection terms,and source/sink terms.Two methods are developed to quantify the CWR(details of which are presented in the companion paper)with satellite observations,atmospheric reanalysis data,precipitation products,and cloud resolving models.The CWR and related variables over North China in April and August 2017 are thus derived.The results show that CWR has the same order of magnitude as surface precipitation(Ps).The hydrometers converted from water vapor(Cvh)during the condensation process is the primary source of precipitation.It is highly correlated with Ps and contributes the most to the CWR over a large region.The state variables and advection terms of hydrometeors are two orders of magnitude lower than the corresponding terms of water vapor.The atmospheric hydrometeors can lead to higher PE than water vapor(several tens of percent versus a few percent),with a shorter RT(only a few hours versus several days).For daily CWR,the state terms are important,but for monthly and longer-time mean CWR,the source/sink terms(i.e.,cloud microphysical processes)contribute the largest;meanwhile,the advection terms contribute less for larger study areas.

Quantifying the Cloud Water Resource:Methods Based on Observational Diagnosis and Cloud Model Simulation

Based on the concepts of cloud water resource(CWR)and related variables proposed in the first part of this study,this paper provides details of two methods to quantify the CWR.One is diagnostic quantification(CWR-DQ)based on satellite observations,precipitation products,and atmospheric reanalysis data;and the other is numerical quantification(CWR-NQ)based on a cloud resolving model developed at the Chinese Academy of Meteorological Sciences(CAMS).The two methods are applied to quantify the CWR in April and August 2017 over North China,and the results are evaluated against all available observations.Main results are as follows.(1)For the CWR-DQ approach,reference cloud profiles are firstly derived based on the Cloud Sat/CALIPSO joint satellite observations for 2007–2010.The NCEP/NCAR reanalysis data in 2000–2017 are then employed to produce three-dimensional cloud fields.The budget/balance equations of atmospheric water substance are lastly used,together with precipitation observations,to retrieve CWR and related variables.It is found that the distribution and vertical structure of clouds obtained by the diagnostic method are consistent with observations.(2)For the CWR-NQ approach,it assumes that the cloud resolving model is able to describe the cloud microphysical processes completely and precisely,from which four-dimensional distributions of atmospheric water vapor,hydrometeors,and wind fields can be obtained.The data are then employed to quantify the CWR and related terms/quantities.After one-month continuous integration,the mass of atmospheric water substance becomes conserved,and the tempospatial distributions of water vapor,hydrometeors/cloud water,and precipitation are consistent with observations.(3)Diagnostic values of the difference in the transition between hydrometeors and water vapor(Cvh-Chv)and the surface evaporation(Es)are well consistent with their numerical values.(4)Correlation and bias analyses show that the diagnostic CWR contributors are well correlated with observations,and match their numerical counterparts as well,indicating that the CWR-NQ and CWR-DQ methods are reasonable.(5)Underestimation of water vapor converted from hydrometeors(Chv)is a shortcoming of the CWR-DQ method,which may be rectified by numerical quantification results or by use of advanced observations on higher spatiotemporal resolutions.

Diagnostic Quantification of the Cloud Water Resource in China during 2000–2019

By using the diagnostic quantification method for cloud water resource(CWR),the three-dimensional(3D)cloud fields of 1°×1°resolution during 2000-2019 in China are firstly obtained based on the NCEP reanalysis data and related satellite data.Then,combined with the Global Precipitation Climatology Project(GPCP)products,a 1°×1°gridded CWR dataset of China in recent 20 years is established.On this basis,the monthly and annual CWR and related variables in China and its six weather modification operation sub-regions are obtained,and the CWR characteristics in different regions are analyzed finally.The results show that in the past 20 years,the annual total amount of atmospheric hydrometeors(GM_(h))and water vapor(GM_(v))in the Chinese mainland are about 838.1 and 3835.9 mm,respectively.After deducting the annual mean precipitation of China(P_(s),661.7 mm),the annual CWR is about 176.4 mm.Among the six sub-regions,the southeast region has the largest amount of cloud condensation(C_(vh))and precipitation,leading to the largest GM_(h) and CWR there.In contrast,the annual P_(s),GM_(h),and CWR are all the least in the northwest region.Furthermore,the monthly and interannual variation trends of P_(s),C_(vh),and GM_(h) in different regions are identical,and the evolution characteristics of CWR are also consistent with the hydrometeor inflow(Q_(hi)).For the north,northwest,and northeast regions,in spring and autumn the precipitation efficiency of hydrometeors(PEh)is not high(20%-60%),the renewal time of hydrometeors(RT_(h))is relatively long(5-25 h),and GM_(h) is relatively high.Therefore,there is great potential for the development of CWR through artificial precipitation enhancement(APE).For the central region,spring,autumn,and winter are suitable seasons for CWR development.For the southeast and southwest regions,P_(s) and PE_(h) in summer are so high that the development of CWR should be avoided.For different spatial scales,there are significant differences in the characteristics of CWR.

Improving Scalability of Cloud Monitoring Through PCA-Based Clustering of Virtual Machines

Cloud computing has recently emerged as a leading paradigm to allow customers to run their applications in virtualized large-scale data centers. Existing solutions for monitoring and management of these infrastructures consider virtual machines （VMs） as independent entities with their own characteristics. However, these approaches suffer from scalability issues due to the increasing number of VMs in modern cloud data centers. We claim that scalability issues can bc addressed by leveraging the similarity among VMs behavior in terms of resource usage patterns. In this paper we propose an automated methodology to cluster VMs starting from the usage of multiple resources, assuming no knowledge of the services executed on them. The innovative contribution of the proposed methodology is the use of the statistical technique known as principal component analysis （PCA） to automatically select the most relevant information to cluster similar VMs. We apply the methodology to two case studies, a virtualized testbed and a real enterprise data center. In both case studies, the automatic data selection based on PCA allows us to achieve high performance, with a percentage of correctly clustered VMs between 80% and 100% even for short time series （1 day） of monitored data. Furthermore, we estimate the potential reduction in the amount of collected data to demonstrate how our proposal may address the scalability issues related to monitoring and management in cloud computing data centers.

Cloud Water Resource in North China in 2017 Simulated by the CMA-CPEFS Cloud Resolving Model:Validation and Quantification

Based on the concept of cloud water resource(CWR)and the cloud microphysical scheme developed by the Chinese Academy of Meteorological Sciences(CAMS),a coupled mesoscale and cloud-resolving model system is developed in the study for CWR numerical quantification(CWR-NQ)in North China for 2017.The results show that(1)the model system is stable and capable for performing 1-yr continuous simulation with a water budget error of less than 0.2%,which indicates a good water balance.(2)Compared with the observational data,it is confirmed that the simulating capability of the CWR-NQ approach is decent for the spatial distribution of yearly cumulative precipitation,daily precipitation intensity,yearly average spatial distribution of water vapor.(3)Compared with the CWR diagnostic quantification(CWR-DQ),the results from the CWR-NQ differ mainly in cloud condensation and cloud evaporation.However,the deviation of the net condensation(condensation minus evaporation)between the two methods is less than 1%.For other composition variables,such as water vapor advection,surface evaporation,precipitation,cloud condensation,and total atmospheric water substances,the relative differences between the CWR-NQ and the CWR-DQ are less than 5%.(4)The spatiotemporal features of the CWR in North China are also studied.The positive correlation between water vapor convergence and precipitation on monthly and seasonal scales,and the lag of precipitation relative to water vapor convergence on hourly and daily scales are analyzed in detail,indicating the significance of the state term on hourly and daily scales.The effects of different spatial scales on the state term,advection term,source-sink term,and total amount are analyzed.It is shown that the advective term varies greatly at different spatiotemporal scales,which leads to differences at different spatiotemporal scales in CWR and related characteristic quantities.

A Study on the Cloud Design Resource System Based on Ecology Theory

The cloud design resource system is an organic entity from the ecological system perspective, whose elements have the similar behavioral property of creature individuals, populations, and community. Hence, study on cloud design resource management can refer to the ecology theory. Based on analyzing the structure of cloud design resource system and the features of cloud design resources, the similarity between those are further dis cussed through the ecological system elements and features. An ecological based application framework of cloud design resource management is then presented.

An Optimal Resource Provision Policy in Cloud Computing Based on Customer Profiles

Conventional resource provision algorithms focus on how to maximize resource utilization and meet a fixed constraint of response time which be written in service level agreement（SLA）.Unfortunately,the expected response time is highly variable and it is usually longer than the value of SLA.So,it leads to a poor resource utilization and unnecessary servers migration.We develop a framework for customer-driven dynamic resource allocation in cloud computing.Termed CDSMS（customer-driven service manage system）,and the framework’s contributions are twofold.First,it can reduce the total migration times by adjusting the value of parameters of response time dynamically according to customers’profiles.Second,it can choose a best resource provision algorithm automatically in different scenarios to improve resource utilization.Finally,we perform a serious experiment in a real cloud computing platform.Experimental results show that CDSMS provides a satisfactory solution for the prediction of expected response time and the interval period between two tasks and reduce the total resource usage cost.