Matchmaking with limited knowledge of resources on clouds and grids

A Survey on Resource Allocation Strategies in Cloud Computing

scholarli z, de souza r, & goh m (2016) supply chain orchestration leveraging on mnc networks and local resources: approach strategies. each node in the network has exponential processing times and a generalized ps policy to approximate the operating system scheduling. scholardantas j, matos r, araujo j, maciel p (2012) an availability model for eucalyptus platform: an analysis of warm-standy replication mechanism. he has published more than 100 papers in various journals and conference proceedings. scholarvakili a & navimipour nj (2017) comprehensive and systematic review of the service composition mechanisms in the cloud environments. scholargajbhiye a & shrivastva kmp (2014) cloud computing: need, enabling technology, architecture, advantages and challenges. jhawar and piuri [74] uses fault trees and markov models to evaluate the reliability and availability of a cloud system under different deployment contexts. [118] introduces a client-side admission control method to schedule requests among vms, looking at minimizing the cost of application, sla violations and iaas resources. queueing theory is used to model different components of the system and data mining and machine learning approaches ensure dynamic adaptation of the model to work under system fluctuations. hardware heterogeneity and vm interference are the primary cause for such variability, which is also visible within vms of the same instance class. complications arise in workload inference on paas clouds, where infrastructure-level metrics such as cpu utilization are normally unavailable to the users. scholarkeshanchi b, souri a, & navimipour nj (2017) an improved genetic algorithm for task scheduling in the cloud environments using the priority queues: formal verification, simulation, and statistical testing. in: proceedings of the 4th usenix conference on hot topics in cloud ccomputing, hotcloud’12, 4–4, boston, ma, usagoogle scholarschad j, dittrich j, quiané-ruiz j-a: runtime measurements in the cloud: observing, analyzing, and reducing variance. the model accounts for time variations in vm resource usage, and it is used as the basis of a vm consolidation algorithm. in: proceedings of the 2013 ieee/ifip 43rd international conference on dependable systems and networks, dsn 2013, 1–6, hong kong, china. mao and humphrey [111] defines an auto-scaling mechanism to guarantee the execution of all jobs within given deadlines. scholarpadala p, shin kg, zhu x, uysal m, wang z, singhal s, merchant a, salem k (2007) adaptive control of virtualized resources in utility computing environments. the proposed algorithms consider customer profiles and quality parameters to cope with dynamic workloads and heterogeneous cloud resources. scholarcalheiros rn, netto mas, de rose caf, buyya r: emusim: an integrated emulation and simulation environment for modeling, evaluation, validation of performance of cloud computing applications. [22] applies pattern recognition techniques to data center and cloud workload data. scholarcalheiros rn, buyya r, de rose caf: building an automated and self-configurable emulation testbed for grid applications. in artificial intelligence and evolutionary algorithms in engineering systems, ed: springer:341–348. a comparison of the pros and cons of some popular stochastic formalisms can be found in [46], where the authors highlight the issue that a given method can perform better on some system model but not on others, making it difficult to make absolute recommendations on the best model to use. scholarsrinivasan s (2014) security, trust, and regulatory aspects of cloud computing in business environments:igi global. in: proceedings of the 2010 ieee symposium on computers and communications, iscc’10, 695–700, riccione, italy. the authors studied the behavior of both online and offline algorithms by means of a simulation campaign. scholarwei b, lin c, kong x (2011) dependability modeling and analysis for the virtual data center of cloud computing. surprisingly, we have found a limited amount of work specific to workload analysis and inference techniques in the cloud. scholarrolia j, vetland v: parameter estimation for performance models of distributed application systems. scholarmenascé d, almeida v, dowdy l: capacity planning and performance modeling: from mainframes to client-server systems. we also discuss some techniques originally developed for modeling and dynamic management in enterprise data centers that have been successively applied in the cloud context. we review in section 2 cloud measurement studies that help characterize those properties for specific cloud systems, followed by a review of workload characterizations and inference techniques that can be applied to qos analysis. these include, among others, difficulties in modeling caching, lack of methods to compute percentiles of response times, tradeoff between accuracy and speed. scholarghosh s & me dr (2014) enhanced distributed accountability framework with indexing in the cloud and its security analysis. scholaralamir p, jafari navimipour n, ramage m, ramage m, & ramage m (2016) trust evaluation between the users of social networks using the quality of service requirements and call log histories.., reliability block diagrams and fault trees, for analyzing mean time to failure (mttf) and mean time between failures (mtbf). load balancer is presented in [50] to assign vms among geographically-distributed data centers considering predictions on workload, energy prices, and renewable energy generation capacities. indeed, a trade-off exists between available information, qos model complexity, computational cost of decision-making, and accuracy of predictions. gspns are used to provide fine-grained detail on the inner vm behaviors, such as separation of privileged and non-privileged instructions and successive handling by the vm or the vm monitor. in: proceedings of the 2012 ieee international conference on systems, man, and cybernetics, smc 2012, 1664–1669, seoul, korea. scholartakabi h, joshi jb, & ahn gj (2010) security and privacy challenges in cloud computing environments. scholarzhang q, cheng l, boutaba r: cloud computing: state-of-the-art and research challenges. bonvin n, papaioannou t, aberer k (2010) an economic approach for scalable and highly-available distributed applications.. cloud computing is an operation model that integrates many technological advancements of the last decade such as virtualization, web services, and sla management for enterprise applications. while some bias components can be filtered out (for example using the cpu steal metric available on amazon ec2 virtual machines), contention on resources such as cache, memory bandwidth, network, or storage, is harder or even impossible to monitor for the final user. the load balancer uses the number of outstanding requests and the inter-departure times in each vm to dispatch requests to the vm with the shortest expected response time. then the qos-aware service composition is solved by influence diagrams followed by analytical and simulation experiments. the aim of the paper is to exploit resources in domains with low allocation costs and, at the same time, achieve better network performance among cloud nodes. scholartakabi h, joshi jb, & ahn gj (2010) security and privacy challenges in cloud computing environments. the consolidation algorithm is tested and shown to be highly competitive. if a component fails, it assumes the logical value true, and the failure propagation can be studied via the tree structure. a refinement process is conducted between clustering and regression to get accurate clustering results by removing outliers and merging the clusters that fit the same model. then, the problem to be addressed is to determine the minimum number of vms or containers needed to fulfill the target qos, pursuing the best trade-off between cost and performance. scholarcalinescu r, ghezzi c, kwiatkowska mz, mirandola r: self-adaptive software needs quantitative verification at runtime.

New Roadmap for Elastic Grid Resource Matchmaking

scholarsun d, chang g, sun l, & wang x (2011) surveying and analyzing security, privacy and trust issues in cloud computing environments. the cloud technology stack has also become mainstream in enterprise data centers, where private and hybrid cloud architectures are increasingly adopted. in: proceedings of the 2013 ieee 5th international conference on cloud computing technology and science, volume 1 of cloudcom 2013, 1–8, bristol, united kingdom. in: proceedings of the 2011 ieee ninth international conference on dependable, autonomic and secure computing, dasc 2011, 449–456, sydney, nsw, australiagoogle scholarardagna d, pernici b: adaptive service composition in flexible processes. scholaravram mg (2014) advantages and challenges of adopting cloud computing from an enterprise perspective. the algorithm evaluates allocation and revenues as the users place requests to the system. scholarjanuzaj y, ajdari j, & selimi b (2015) dbms as a cloud service: advantages and disadvantages. in: proceedings of the 6th international symposium on software engineering for adaptive and self-managing systems, seams ’11, 218–227, honolulu, hi, usa. petri nets have been extended to consider stochastic transitions, in stochastic petri nets (spns) and generalized spns (gspns). the authors demonstrate that their solution is able to improve the variance and percentiles of response times with respect to a built-in policy of the apache web server. [123] a structured peer-to-peer network, based on distributed hash tables, is proposed to support service discovery, self-management, and load-balancing of cloud applications. a comparison against a set of heuristics from the literature and an oracle with perfect knowledge about the future load shows that the proposed algorithm overcomes the heuristic approaches, without penalizing slas and it is able to produce results that are close to the global optimum. learning techniques, instead, use learning mechanisms to capture the behavior of the system without any explicit performance or traffic model and with little built-in system knowledge. in particular, the m/g/1 ps queue is a common abstraction used to model a cpu and it has been adopted in many cloud studies [47],[48], thanks to its simplicity and the suitability to apply the model to multi-class workloads..001google scholarcremonesi p, sansottera a: indirect estimation of service demands in the presence of structural changes. qos properties have received constant attention well before the advent of cloud computing, performance heterogeneity and resource isolation mechanisms of cloud platforms have significantly complicated qos analysis, prediction, and assurance. from the perspective of cloud providers and users, inference techniques provide a means to estimate the workload profile of individual vms running on their infrastructures, taking into account hidden variables due to lack of information. scholarshahzad f (2014) state-of-the-art survey on cloud computing security challenges, approaches and solutions. scholarsedaghat m, hernandez-rodriguez f, elmroth e (2013) a virtual machine re-packing approach to the horizontal vs. important issues are treated, such as the selection of the most relevant data, the modeling technique, and variable-selection procedure. jhawar and piuri [74] uses fault trees and markov models to evaluate the reliability and availability of a cloud system under different deployment contexts. [37] considers the problem of dynamically estimating cpu demands of diverse types of requests using cpu utilization and throughput measurements. scholarcopyright information© the society of service science and springer-verlag gmbh germany 2017authors and affiliationsmatin chiregi1nima jafari navimipour2email author1. [35] proposes the demand estimation with confidence (dec) approach to overcome the problem of multicollinearity in regression methods. lqns are here useful to handle the complexity of geo-distributed applications that include both transactional and streaming workloads. [129] also considers the vm consolidation problem by modeling the vm resource demands as a set of correlated random variables. spns provide a convenient way in this setting to represent energy flow and cooling in the infrastructure. [43] studies the relations between workload and resource consumption for cloud web applications. qos is fundamental for cloud users, who expect providers to deliver the advertised quality characteristics, and for cloud providers, who need to find the right tradeoffs between qos levels and operational costs. many cloud operators are now active on the market, providing a rich offering, including infrastructure-as-a-service (iaas), platform-as-a-service (paas), and software-as-a-service (saas) solutions [2]. heterogeneity in customer slas is handled in [52] with an m/m/k/k priority queue, which is a queue with exponentially distributed inter-arrival times and service times, k servers and no buffer. one of the challenges posed by cloud applications is quality-of-service (qos) management, which is the problem of allocating resources to the application to guarantee a service level along dimensions such as performance, availability and reliability.., response time, throughput and resource utilization) predicted by a performance model against measurements collected in a controlled experimental environment.., reliability block diagrams and fault trees, for analyzing mean time to failure (mttf) and mean time between failures (mtbf). in artificial intelligence and evolutionary algorithms in engineering systems, ed: springer:341–348. his research interests include cloud computing, social networks, fault-tolerance software, computational intelligence, evolutionary computing, and network on chip.) vice versa, integrating workload characterisation, system models and resource management solutions, pro-active systems, may help to prevent qos degradation..284mathscinetgoogle scholarzaman s, grosu d (2012) an online mechanism for dynamic vm provisioning and allocation in clouds. regression techniques can exploit these formulas to obtain demand estimates from system measurements [29]-[33]. a comparison of the pros and cons of some popular stochastic formalisms can be found in [46], where the authors highlight the issue that a given method can perform better on some system model but not on others, making it difficult to make absolute recommendations on the best model to use. the solution methods are based on the bender decomposition approach to divide the problem into sub-problems, which can be solved in parallel, and an approximation algorithm to solve problems with a large set of scenarios. fault trees and markov models are used to evaluate the reliability and availability of fault tolerance mechanisms. an analytical model, based on queueing theory, is presented to describe the relation between the number of replicas and the service level, e. finally, the presented approach is endorsed against fixed and adaptive control schemes by means of a campaign of experiments. here we briefly review queueing models, petri nets, and other specialized formalisms for reliability evaluation. [100] is a simulator for scientific applications deployed on large-scale clouds and grids. in 2010 international conference on computer application and system modeling (iccasm 2010). in bio-inspired models of network, information, and computing systems, volume 87 of lecture notes of the institute for computer sciences, social informatics and telecommunications engineering. ford b (2012) icebergs in the clouds: the other risks of cloud computing. scholarsutton ca, jordan mi (2010) inference and learning in networks of queues. in proceedings of the 2008 acm sigmetrics international conference on measurement and modeling of computer systems. a lot of work has been done in the last decade for optimal admission control in web servers and multi-tier applications. we also review and classify their early application to some decision-making problems arising in cloud qos management. and fault trees aim at obtaining the overall system reliability from the reliability of the system components.

  • A Semantic Framework for Resource Discovery Based on Ontology

    the pattern and trend are first analyzed and then synthetic workloads are created to reflect future behaviors of the workload. scholartchifilionova v (2011) security and privacy implications of cloud computing-lost in the cloud. [34] presents an optimization-based inference technique that is formulated as a robust linear regression problem that can be used with both closed and open queueing network performance models. nevertheless, training sessions tend to extend over several hours [144] and retraining is required for evolving workloads. in: proceedings of the 2011 ieee international conference on systems, man, and cybernetics, smc 2011, 421–426, anchorage, ak, usa. inference is often justified by the overheads of deep monitoring and by the difficulty of tracking execution paths of individual requests [25]. theory has the advantage of guaranteeing the stability of the system upon workload changes by modeling the transient behavior and adjusting system configurations within a transitory period [143]. methods are considered in [107], which proposes a self-organizing approach to provide robust and scalable solutions for service deployment, resource provisioning, and load balancing in a cloud infrastructure. [52] allows service providers to reserve a certain amount of resources exclusively for some customers, according to slas. the knowledge of request flow intensities provides throughputs that can be used in regression techniques. for example, greencloud [97], which is an extension of the packet-level simulator ns2 [98], aims at evaluating the energy consumption of the data center resources where the application has been deployed, considering servers, links, and switches. many cloud operators are now active on the market, providing a rich offering, including infrastructure-as-a-service (iaas), platform-as-a-service (paas), and software-as-a-service (saas) solutions [2]. different admission control and scheduling algorithms are proposed in [119] to effectively exploiting public cloud resources. models have been used primarily in optimising web service composition [76], but they are now becoming relevant also in the description of saas applications, iaas resource orchestration, and cloud-based business-process execution. hence, in this paper, the comprehensive and detailed study and survey of the state of the art techniques and mechanisms in this field are provided. scholarsutton ca, jordan mi (2010) inference and learning in networks of queues. scholarwei b, lin c, kong x (2011) dependability modeling and analysis for the virtual data center of cloud computing. scholarkeshanchi b, souri a, & navimipour nj (2017) an improved genetic algorithm for task scheduling in the cloud environments using the priority queues: formal verification, simulation, and statistical testing. the solution provides an effective decentralized algorithm for deploying massively scalable services and it is suitable for all the situations in which a centralized solution is not feasible. the work uses a probabilistic approach to find an optimized allocation of services on virtualized physical resources. also, we discuss the trust evaluating mechanisms in the cloud computing and categorize them into two main groups including centralized and distributed mechanisms. variants of these regression methods have been developed to cope with problems such as outliers [34], data multi-collinearity [35], online estimation [36], data aging [37], handling of multiple system configurations [38], and automatic definition of request types [39],[40]. the authors present a network model that allows estimating latencies between locations and propose a genetic algorithm to achieve network-aware and qos-aware service provisioning. in: proceedings of the 2011 ieee ninth international conference on dependable, autonomic and secure computing, dasc 2011, 449–456, sydney, nsw, australiagoogle scholarardagna d, pernici b: adaptive service composition in flexible processes., [105] proposes an adaptive approach for component replication of cloud applications, aiming at finding a cost-effective placement and load balancing. [18] proposes a model-predictive resource allocation algorithm that auto-scales vms, with the aim of optimizing the utility of the application over a limited prediction horizon..001google scholarcremonesi p, sansottera a: indirect estimation of service demands in the presence of structural changes. in: proceedings of the 2013 acm cloud and autonomic computing conference, cac ’13, 6:1–6:10, miami, fl, usa,google scholarali-eldin a, tordsson j, elmroth e (2012) an adaptive hybrid elasticity controller for cloud infrastructures. we consider the resource management mechanisms for applications qos enforcement provided by public clouds, they are quite simplistic if compared to current research proposals. the authors formulate an optimization problem faced by a cloud procurement endpoint (a module responsible for provisioning resources from public cloud providers), where heavy workloads are tackled by relying on public clouds. this may be helpful, for instance, to jointly tackle capacity allocation and load balancing. [52] allows service providers to reserve a certain amount of resources exclusively for some customers, according to slas. scholarjhawar r, piuri v (2012) fault tolerance management in iaas clouds. however, description of performance is often quite basic and associated with mean resource requirements of the applications. scholargasquet c, witomski p: fourier analysis and applications: filtering, numerical computation, wavelets, volume 30 of texts in applied mathematics. emulation is used to understand the application behavior, extracting profiling information. the authors use this model to investigate rejection probabilities and help dimensioning of cloud data centers.., when part of the application traffic is redirected from a private to a public data center to cope with a traffic intensity that surpasses the capacity of the private infrastructure), if the public cloud resources are not provided timely, one can decide to drop new incoming request to preserve the qos for users already in the system (or at least part of them, e. scholarpatel n & chauhan s (2015) a survey on load balancing and scheduling in cloud computing. most of the techniques used for traffic forecasting, resource consumption estimation, and anomaly detection have received little or no validation in a cloud environment. several works instead adopt a description that includes standard deviations [76],[82],[83] or finite ranges of variability for the execution times [84],[85]. in: proceedings of 2012 international symposium on cloud and services computing, iscos 2012, 25–30, mangalore, india. a comparison with state of the art qos routing algorithms shows that the proposed algorithm is both cost-effective and lightweight. scholardrago i, mellia m, munafo mm, sperotto a, sadre r, pras a (2012) inside dropbox: understanding personal cloud storage services. in: proceedings of the 2010 24th ieee international conference on advanced information networking and applications, aina 2010, 446–452, perth, australia. complications arise in workload inference on paas clouds, where infrastructure-level metrics such as cpu utilization are normally unavailable to the users. scholarstewart c, chakrabarti a, griffith r (2013) zoolander: efficiently meeting very strict, low-latency slos. in: proceedings of the 7th international conference on network and services management, 1–9. an evaluation of regression techniques least squares (lsq), least absolute deviations (lad) and support vector regression (svr) is presented. anselmi and casale [120] provides a simple heuristic for user-side load-balancing under connection pooling that is validated against an iaas cloud dataset. a common workload inference approach involves estimating only the mean demand placed by a given type of requests on the resource [26]-[28]. learning techniques, instead, use learning mechanisms to capture the behavior of the system without any explicit performance or traffic model and with little built-in system knowledge..21google scholarbacigalupo d, van hemert j, chen x, usmani a, chester a, he l dillenberger d, wills g, gilbert l, jarvis s: managing dynamic enterprise and urgent workloads on clouds using layered queuing and historical performance models. [75] presents a methodology to identify, mitigate, and monitor risks in cloud resource provisioning.
  • QoS-aware and Semantic-based Service Coordination for Multi

    the model accounts for time variations in vm resource usage, and it is used as the basis of a vm consolidation algorithm. scholarkhazaei h, misic j, misic v, rashwand s: analysis of a pool management scheme for cloud computing centers. in: proceedings of the 2011 ieee 13th international conference on high performance computing and communications, hpcc 2011, 784–789, bamff, canada. scholarbeloglazov a, buyya r, lee yc, zomaya ay: a taxonomy and survey of energy-efficient data centers and cloud computing systems..909764google scholaranandkumar a, bisdikian c, agrawal d: tracking in a spaghetti bowl: monitoring transactions using footprints. the solution methods are based on the bender decomposition approach to divide the problem into sub-problems, which can be solved in parallel, and an approximation algorithm to solve problems with a large set of scenarios. [70], the authors propose a methodology to evaluate data center power infrastructures considering both reliability and cost. in: proceedings of the 10th usenix conference on networked systems design and implementation, nsdi ’13, 329–342, lombard, il, usa. in: proceedings of the 2011 ieee international conference on systems, man, and cybernetics, smc 2011, 421–426, anchorage, ak, usa. in: proceedings of the 2013 15th international symposium on symbolic and numeric algorithms for scientific computing, synasc ’13, 409–416, timisoara, romaniagoogle scholarfaisal a, petriu d, woodside m (2013) network latency impact on performance of software deployed across multiple clouds. the proposed framework helps to stipulate a realistic sla with customers and supports dynamic load shedding and capacity provisioning by considering a queueing model with multiple priority classes. different admission control and scheduling algorithms are proposed in [119] to effectively exploiting public cloud resources. the cloud-user perspective, the admission control mechanism is used as an extreme overload mechanism, helpful when additional resources are obtained with some significant delay. this paper aims at supporting these efforts by providing a survey of the state of the art of qos modeling approaches applicable to cloud computing and by describing their initial application to cloud resource management. of service science researchjune 2017, volume 9, issue 1,Pp 1–30 | cite asa comprehensive study of the trust evaluation mechanisms in the cloud computingauthorsauthors and affiliationsmatin chireginima jafari navimipouremail authorresearch papersfirst online: 30 june 2017received: 28 october 2015accepted: 19 march 2017. common analytical formulas involve queues with exponential service and arrival times, with a single server (m/m/1) or with k servers (m/m/k), and queues with generally-distributed service times (m/g/1). scholardantas j, matos r, araujo j, maciel p (2012) an availability model for eucalyptus platform: an analysis of warm-standy replication mechanism. nets, reliability block diagrams (rbd), and fault trees are probably the most widely known and used formalisms for dependability analysis. scholarxie x, liu r, cheng x, hu x, & ni j (2016) trust-driven and pso-sfla based job scheduling algorithm on cloud. petri nets are a flexible and expressive modeling approach, which allows a general interactions between system components, including synchronization of event firing times. [21] defines a bayesian algorithm for long-term workload prediction and pattern analysis, validating results on data from a google data center. in: proceedings of the 2012 ieee network operations and management symposium, noms 2012, 1287–1294, maui, hi, usa. while queueing systems are widely used to model single resources subject to contention, queueing networks are able to capture the interaction among a number of resources and/or applications components. queueing models use the knowledge of the system topology and infrastructure to provide accurate performance predictions. scholarcalheiros rn, ranjan r, beloglazov a, de rose caf, buyya r: cloudsim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms. in fact, the probability of requesting more resources for a service is at the basis of the formulated optimization model, that constitutes a probabilistic admission control test. computing is a new model to enable convenient and on-demand access to the pool of configurable computing resources. [69] proposes the use of gspns to evaluate the impact of virtualization mechanisms, such as vm consolidation and live migration, on cloud infrastructure dependability. scholarcopyright information© the society of service science and springer-verlag gmbh germany 2017authors and affiliationsmatin chiregi1nima jafari navimipour2email author1. scholardrago i, mellia m, munafo mm, sperotto a, sadre r, pras a (2012) inside dropbox: understanding personal cloud storage services. important issues are treated, such as the selection of the most relevant data, the modeling technique, and variable-selection procedure. modeling involves the assessment or prediction of the arrival rates of requests and of the demand for resources (e. scholarvan den, bossche r, vanmechelen k, broeckhove j (2010) cost-optimal scheduling in hybrid iaas clouds for deadline constrained workloads. scholarranjan r, zhao l, wu x, liu a, quiroz a, parashar m: peer-to-peer cloud provisioning: service discovery and load-balancing. scholarmoyano f, fernandez-gago c, & lopez j (2013) a framework for enabling trust requirements in social cloud applications. forecasting and trend analysis techniques are commonly used to predict web traffic intensity at different timescales. scholarghosh s & me dr (2014) enhanced distributed accountability framework with indexing in the cloud and its security analysis. heterogeneity in customer slas is handled in [52] with an m/m/k/k priority queue, which is a queue with exponentially distributed inter-arrival times and service times, k servers and no buffer. applications to cloud qos modeling include the use of spns to evaluate the dependability of a cloud infrastructure [68], considering both reliability and availability. scholargmach d, rolia j, cherkasova l, kemper a (2007) workload analysis and demand prediction of enterprise data center applications. the solution provides an effective decentralized algorithm for deploying massively scalable services and it is suitable for all the situations in which a centralized solution is not feasible. scholarkusic d, kandasamy n (2006) risk-aware limited lookahead control for dynamic resource provisioning in enterprise computing systems. of service science researchjune 2017, volume 9, issue 1,Pp 1–30 | cite asa comprehensive study of the trust evaluation mechanisms in the cloud computingauthorsauthors and affiliationsmatin chireginima jafari navimipouremail authorresearch papersfirst online: 30 june 2017received: 28 october 2015accepted: 19 march 2017. scholarcalheiros rn, buyya r, de rose caf: building an automated and self-configurable emulation testbed for grid applications. scholarzhang q, zhu q, zhani mf, boutaba r (2012) dynamic service placement in geographically distributed clouds. on the other hand, the analysis also shows that turning on the geographical load balancing has a strong impact on quality of the solutions (between 27% and 40%) of the online algorithm. scholarputhal d, sahoo b, mishra s, & swain s (2015) cloud computing features, issues, and challenges: a big picture.., system throughput and utilization of the servers), commonly retrieved from log files, in order to estimate service times. in: proceedings of the international conference for high performance computing, networking, storage and analysis, sc12, 1–11, salt lake city, utah,usa. emulation is used to understand the application behavior, extracting profiling information. in: proceedings of the 2011 international conference for high performance computing, networking, storage and analysis, sc ’11, 1–12, seattle, wa, usa. moreover, it improves the placement of application instances by putting idle machines into standby mode and reducing the number of running instances in condition of light load. petri nets are a flexible and expressive modeling approach, which allows a general interactions between system components, including synchronization of event firing times. [95] builds on top of cloudsim by adding an emulation step leveraging the automated emulation framework (aef) [96].
  • How often should you see him when first dating
  • Ontology-based Grid resource management

    an evaluation of regression techniques least squares (lsq), least absolute deviations (lad) and support vector regression (svr) is presented. scholarvakili a & navimipour nj (2017) comprehensive and systematic review of the service composition mechanisms in the cloud environments. qos is fundamental for cloud users, who expect providers to deliver the advertised quality characteristics, and for cloud providers, who need to find the right tradeoffs between qos levels and operational costs..74google scholaromari t, franks g, woodside m, pan a: efficient performance models for layered server systems with replicated servers and parallel behaviour. queueing network approach is taken in [66] to provision resources for data-center applications. variants of these regression methods have been developed to cope with problems such as outliers [34], data multi-collinearity [35], online estimation [36], data aging [37], handling of multiple system configurations [38], and automatic definition of request types [39],[40]. scholarpacifici g, segmuller w, spreitzer m, tantawi a: cpu demand for web serving: measurement analysis and dynamic estimation. scholaremeakaroha v, fatema k, vanderwerff l, healy p, lynn t, & morrison j (2016) a trust label system for communicating trust in cloud services. the algorithm developed has the additional benefit to leverage cloud elasticity to allocate and deallocate resources to help services to respect contractual slas. [90] presents a graph-theoretic model for qos-aware service composition in cloud platforms, explicitly handling network virtualization. each node in the network has exponential processing times and a generalized ps policy to approximate the operating system scheduling. cloud applications are often tiered and queueing networks can capture the interactions between tiers. this effectively creates a divide between the knowledge that can be made available for an application by its designers and the techniques used to manage it. queueing models use the knowledge of the system topology and infrastructure to provide accurate performance predictions. techniques have also been used to correlate the cpu demand placed by a request on multiple servers. two complementary methods are proposed: an offline deterministic optimization method to be used at design time and an online vm placement, migration and geographical load balancing algorithm for runtime.., network bandwidth variance, virtual machine (vm) startup times, start failure probabilities. scholarsadashiv n & kumar sd (2011) cluster, grid and cloud computing: a detailed comparison. in: proceedings of the 2013 15th international symposium on symbolic and numeric algorithms for scientific computing, synasc ’13, 409–416, timisoara, romaniagoogle scholarfaisal a, petriu d, woodside m (2013) network latency impact on performance of software deployed across multiple clouds. problem areas covered in this section include capacity allocation, load balancing, and admission control. the proposed algorithm is compared against a greedy heuristic method and it shows significant cost savings (around 20-30%). the system is represented by a set of inter-related blocks, connected by series, parallel, and k-out-of-n relationships. scholarkusic d, kephart jo, hanson je, kandasamy n, jiang g: power and performance management of virtualized computing environments via lookahead control. scholarcalheiros rn, netto mas, de rose caf, buyya r: emusim: an integrated emulation and simulation environment for modeling, evaluation, validation of performance of cloud computing applications. in: proceedings of the 9th usenix conference on file and stroage technologies, fast’11, 12–12, san jose, ca, usa. throughout this paper, we mainly focus on qos aspects pertaining to performance, reliability and availability. in: proceedings of the 2012 ieee international conference on systems, man, and cybernetics, smc 2012, 1664–1669, seoul, korea. [32] presents a queueing network model where each queue represents a tier of a web application, which is parameterized by means of a regression-based approximation of the cpu demand of customer transactions. the drawback is that they adopt a black-box approach, ignoring relevant knowledge of the system that could provide valuable insights into its performance. scholarzhang q, zhu q, zhani mf, boutaba r (2012) dynamic service placement in geographically distributed clouds. here we briefly review queueing models, petri nets, and other specialized formalisms for reliability evaluation. problem areas covered in this section include capacity allocation, load balancing, and admission control. scholarafari navimipour n, rahmani am, navin ah, & hosseinzadeh m (2015) expert cloud: a cloud-based framework to share the knowledge and skills of human resources. decentralized probabilistic algorithm is also described in [106], which focuses on federated clouds. scholarkhethavath p, thomas j, chan-tin e, & liu h (2013) introducing adistributed cloud architecture with efficient resource discovery and optimal resource allocation. in: proceedings of the 9th usenix conference on file and stroage technologies, fast’11, 12–12, san jose, ca, usa. moreover, a closed-form formula for calculating the average response time of a request and a unified framework to manage different levels of slas are provided. throughout this paper, we mainly focus on qos aspects pertaining to performance, reliability and availability. the drawback is that they adopt a black-box approach, ignoring relevant knowledge of the system that could provide valuable insights into its performance..001google scholarhuang j, liu y, duan q (2012) service provisioning in virtualization-based cloud computing: modeling and optimization. the aim of the paper is to exploit resources in domains with low allocation costs and, at the same time, achieve better network performance among cloud nodes. scholarchiregi m & navimipour nj (2016a) a new method for trust and reputation evaluation in the cloud environments using the recommendations of opinion leaders’ entities and removing the effect of troll entities. they use hidden markov models to identify temporal correlations between different clusters and use this information to predict future workload variations. in: proceedings of the 2006 ieee international conference on autonomic computing, icac ’06, 74–83, dublin, ireland. the paper presents a minimum cost maximum flow (mcmf) algorithm and compares it against a modified bin-packing formulation; the mcmf algorithm exhibits very good performance and scalability properties. techniques have also been used to correlate the cpu demand placed by a request on multiple servers. scholarwickremasinghe b, calheiros rn, buyya r (2010) cloudanalyst: a cloudsim-based visual modeller for analysing cloud computing environments and applications.., response times, throughputs, or resource utilizations) to a mean demand to be estimated, e.. the aim of this survey is to provide an overview of early research works in the cloud qos modeling space, categorizing contributions according to relevant areas and methods used. we review in section 2 cloud measurement studies that help characterize those properties for specific cloud systems, followed by a review of workload characterizations and inference techniques that can be applied to qos analysis. this survey covers research efforts in workload modeling, system modeling, and their applications to qos management in the cloud. in: proceedings of the 2013 acm cloud and autonomic computing conference, cac ’13, 6:1–6:10, miami, fl, usa,google scholarali-eldin a, tordsson j, elmroth e (2012) an adaptive hybrid elasticity controller for cloud infrastructures. therefore, one of the most important challenges in this environment is to evaluate the trust value to enable users for selecting the trustworthy resources, however, to the best of our knowledge, the comprehensive and detailed review of the most important techniques in this field is very rare. this survey, a number of insights arise on the current state of the art:The number of works that apply white-box system modeling techniques is quite limited in qos management, albeit popular in the software performance engineering community.
  • Dating advice what to do after the first date
  • Need help with my dating profile
  • Is online dating social networking

Resource Management on Clouds and Grids

Quality-of-service in cloud computing: modeling techniques and

for example, greencloud [97], which is an extension of the packet-level simulator ns2 [98], aims at evaluating the energy consumption of the data center resources where the application has been deployed, considering servers, links, and switches. scholarstewart c, chakrabarti a, griffith r (2013) zoolander: efficiently meeting very strict, low-latency slos. in proceedings of the paper presented at the international conference on image processing, production and computer science, istanbul (turkey). such mechanism uses biological evolution concepts to manage data application services and to produce optimal composition and load balancing solutions. scholarfouladi p & navimipour jn (2017) human resources ranking in a cloud-based knowledge sharing framework using the quality control criteria. the proposed approach is shown to achieve high accuracy for predicting workload and resource usages. several works have investigated over the last two decades the problem of estimating, using indirect measurements, the resource demand placed by an application on physical resources, for example cpu requirements. extensive validation on generated dataset and real data show the effectiveness of the algorithm. scholarzheng t, woodside cm, litoiu m: performance model estimation and tracking using optimal filters. the performance models are based on queueing-network models abstracted from the system and enhanced by machine learning algorithms to correlate system workload attributes with performance attributes. scholarwang y, chandrasekhar s, singhal m, & ma j (2016) a limited-trust capacity model for mitigating threats of internal malicious services in cloud computing. the objectives are the minimization of both resource and penalty costs, as well as minimizing sla violations..23google scholarkalbasi a, krishnamurthy d, rolia j, richter m (2011) mode: mix driven on-line resource demand estimation. work in [139] proposes an admission control protocol to prevent over-utilization of system resources, classifying applications based on resource quality requirements. and conclusionin recent years, cloud computing has matured from an early-stage solution to a mainstream operational model for enterprise applications. in proceedings of the 1st international workshop on software and performance. also, we discuss the trust evaluating mechanisms in the cloud computing and categorize them into two main groups including centralized and distributed mechanisms. scholarzhang q, cheng l, boutaba r: cloud computing: state-of-the-art and research challenges. models have been used primarily in optimising web service composition [76], but they are now becoming relevant also in the description of saas applications, iaas resource orchestration, and cloud-based business-process execution. challenges a threat to workload inference on iaas clouds is posed by resource contention by other users, which can systematically result in biased readings of performance metrics. the main result is that the presented approach is able to provide tight guarantees on the optimality gap and experimental results show that it is at the same time accurate and fast. also, [56] uses a queueing network to represent a multi-tier application deployed in a cloud platform, and to derive an sla-aware resource allocation policy. scholarxie x, liu r, cheng x, hu x, & ni j (2016) trust-driven and pso-sfla based job scheduling algorithm on cloud. [20] uses hidden markov models to capture and predict temporal correlations between workloads of different compute clusters in the cloud.) vice versa, integrating workload characterisation, system models and resource management solutions, pro-active systems, may help to prevent qos degradation. in: proceedings of the 10th usenix conference on networked systems design and implementation, nsdi ’13, 329–342, lombard, il, usa. the knowledge of request flow intensities provides throughputs that can be used in regression techniques. given the lack of control over the system workload and configuration during operation, techniques of this type may not be applicable to production systems for online model calibration. we also discuss some techniques originally developed for modeling and dynamic management in enterprise data centers that have been successively applied in the cloud context. this survey covers research efforts in workload modeling, system modeling, and their applications to qos management in the cloud. the algorithm reduces resources under- and over-provisioning by minimizing the total cost for a customer during a certain time horizon. also, several works have shown how combining the queueing theoretic formulas used by regression methods with the kalman filter can enable continuous demand tracking [41],[42]. in cloud computing, fault trees have been used to evaluate dependencies of cloud services and their effect on application reliability [73]. the authors reduce costs considering the balance of multi-dimensional resources utilization and building up an optimization method for resource allocation; as far as reconfiguration is concerned, they propose a strategy for vm adjustment based on time-division multiplex and on vm live migration. in this paper, the authors propose a method to characterize and predict workloads in cloud environments in order to efficiently provision cloud resources. however, description of performance is often quite basic and associated with mean resource requirements of the applications. scholarkhan a, yan x, shu t, anerousis n (2012) workload characterization and prediction in the cloud: a multiple time series approach. this is prompting several researchers to investigate automated qos management methods that can leverage the high programmability of hardware and software resources in the cloud [4]. scholarnavimipour nj (2015) a formal approach for the specification and verification of a trustworthy human resource discovery mechanism in the expert cloud. cloudanalyst [94] is an extension of cloudsim that allows the modeling of geographically-distributed workloads served by applications deployed on a number of virtualized data centers. load balancer is presented in [50] to assign vms among geographically-distributed data centers considering predictions on workload, energy prices, and renewable energy generation capacities. furthermore, we defined trust characteristics such as integrity, security, availability, reliability, dependability, safety, dynamicity, confidentiality and scalability, and we discuss the trust applications including monitoring and tracking. compared to ordinary queueing networks, lqns provide the ability to describe dependencies arising in a complex workflow of requests and the layering among hardware and software resources that process them. the authors also provide a discussion about the pros and cons of lqns identifying a number of key limitations for their practical use in cloud systems. scholarjuan d & zheng q (2014) cloud and open bim-based building information interoperability research. it first uses nonlinear regression to predict the probability for a query to meet its requirement, and then decides whether the query should be admitted to the database system or not. this approach also considers migration costs, and the multi-dimensional nature of the vm resource requirements (e. spns provide a convenient way in this setting to represent energy flow and cooling in the infrastructure.-demand and reserved resources are considered in the model proposed in [107] to define a bio-inspired self-adapting solution for cloud resource provisioning with the aim of minimizing the number of required virtual machines while meeting slas. in: proceedings of the 2013 13th ieee/acm international symposium on cluster, cloud and grid computing, ccgrid 2013, 327–334, delft, nederlandsgoogle scholarli a, yang x, kandula s, zhang m (2010) cloudcmp: comparing public cloud providers. the cloud technology stack has also become mainstream in enterprise data centers, where private and hybrid cloud architectures are increasingly adopted..909764google scholaranandkumar a, bisdikian c, agrawal d: tracking in a spaghetti bowl: monitoring transactions using footprints. some studies on amazon ec2 have found high-performance contention in cpu-bound jobs [9] and network performance overheads [10]. however, finding optimal tradeoff is a difficult decision problem, often exacerbated by the presence of service level agreements (slas) specifying qos targets and economical penalties associated to sla violations [3].

A comprehensive study of the trust evaluation mechanisms in the

the authors present a network model that allows estimating latencies between locations and propose a genetic algorithm to achieve network-aware and qos-aware service provisioning. scholarjanuzaj y, ajdari j, & selimi b (2015) dbms as a cloud service: advantages and disadvantages. in: proceedings of the international conference for high performance computing, networking, storage and analysis, sc12, 1–11, salt lake city, utah,usa. this survey, a number of insights arise on the current state of the art:The number of works that apply white-box system modeling techniques is quite limited in qos management, albeit popular in the software performance engineering community. in fact, the probability of requesting more resources for a service is at the basis of the formulated optimization model, that constitutes a probabilistic admission control test. scholarsrinivasan s (2014) security, trust, and regulatory aspects of cloud computing in business environments:igi global. scholarel haddad j, manouvrier m, rukoz m: tqos: transactional and qos-aware selection algorithm for automatic web service composition.., from the cloud provider perspective) and resource management techniques for the infrastructure user (e. scholarko rk, jagadpramana p, mowbray m, pearson s, kirchberg m, liang q (2011) trustcloud: a framework for accountability and trust in cloud computing. scholaravram mg (2014) advantages and challenges of adopting cloud computing from an enterprise perspective. the paper takes the perspective of a saas provider with the aim of maximizing the profit by minimizing cost and improving customer satisfaction levels. in this paper, the authors propose a method to characterize and predict workloads in cloud environments in order to efficiently provision cloud resources. scholarleitner p, hummer w, satzger b, inzinger c, dustdar s (2012) cost-efficient and application sla-aware client side request scheduling in an infrastructure-as-a-service cloud. in computational intelligence and networks (cine), 2015 international conference on:116-123. scholarli x, ma h, zhou f, & gui x (2015) service operator-aware trust scheme for resource matchmaking across multiple clouds. there is a growing interest towards understanding better cloud spot markets, where bidding strategies are developed for procuring computing resources. the cloud-user perspective, the admission control mechanism is used as an extreme overload mechanism, helpful when additional resources are obtained with some significant delay. regression techniques can exploit these formulas to obtain demand estimates from system measurements [29]-[33]..066google scholarfranks g, al-omari t, woodside cm, das o, derisavi s: enhanced modeling and solution of layered queueing networks. ford b (2012) icebergs in the clouds: the other risks of cloud computing. policies differ for the decision approach and for the amount of information they use. an extensible meta-model and a class library with an initial set of five models are developed. most of the techniques used for traffic forecasting, resource consumption estimation, and anomaly detection have received little or no validation in a cloud environment. each host can run several vms, and has a power model to determine the overall data center power consumption. scholarsowmya k, sundarraj rp (2012) strategic bidding for cloud resources under dynamic pricing schemes. scholaragostinho l, feliciano g, olivi l, cardozo e, guimaraes e (2011) a bio-inspired approach to provisioning of virtual resources in federated clouds. qos denotes the levels of performance, reliability, and availability offered by an application and by the platform or infrastructure that hosts ita. scholarmao m, humphrey m (2011) auto-scaling to minimize cost and meet application deadlines in cloud workflows. [123] a structured peer-to-peer network, based on distributed hash tables, is proposed to support service discovery, self-management, and load-balancing of cloud applications. [22] applies pattern recognition techniques to data center and cloud workload data. however, the diversity of technologies used in cloud systems makes it difficult to analyze their qos and, from the provider perspective, to offer service-level guarantees. methods are considered in [107], which proposes a self-organizing approach to provide robust and scalable solutions for service deployment, resource provisioning, and load balancing in a cloud infrastructure. here, we survey workload characterization studies and related modeling techniques. research work has focused on policies that are either simple to implement, and thus minimize overheads, or that offer some optimality guarantees, typically proven by analytical models. the pattern and trend are first analyzed and then synthetic workloads are created to reflect future behaviors of the workload. [75] presents a methodology to identify, mitigate, and monitor risks in cloud resource provisioning. and conclusionin recent years, cloud computing has matured from an early-stage solution to a mainstream operational model for enterprise applications. [124] optimizes the allocation and scheduling of vms in federated clouds using a genetic algorithm. an autoregressive process is used to predict the fluctuating incoming demand. a non-linear model for the capacity allocation and load redirection of multiple request classes is proposed and solved by decomposition. we survey in section 3 formalisms and tools employed for these analyses and their current applications to assess the performance of cloud systems..001google scholarhuang j, liu y, duan q (2012) service provisioning in virtualization-based cloud computing: modeling and optimization. farley b, juels a, varadarajan v, ristenpart t, bowers kd, swift mm (2012) more for your money: exploiting performance heterogeneity in public clouds. scholarkhethavath p, thomas j, chan-tin e, & liu h (2013) introducing adistributed cloud architecture with efficient resource discovery and optimal resource allocation. [130] develops an analytical model for resource provisioning, virtual machine deployment, and pool management. scholarsedaghat m, hernandez-rodriguez f, elmroth e (2013) a virtual machine re-packing approach to the horizontal vs.., cpu requirements) placed by applications on an infrastructure or platform, and the qos observed in response to such workloads. capacity allocation problem in presented in [113] that exploits both horizontal and vertical elasticity. scholaranselmi, j, ardagna d, & passacantando m (2014) generalized nash equilibria for saas/ paas clouds. techniques to determine optimized decisions range from simple heuristics to nonlinear programming and meta-heuristics. cloud computing centralized distributed monitoring and tracking matin chiregi received her b.., response times, throughputs, or resource utilizations) to a mean demand to be estimated, e. the paper takes the perspective of a saas provider with the aim of maximizing the profit by minimizing cost and improving customer satisfaction levels. scholarmoyano f, fernandez-gago c, & lopez j (2013) a framework for enabling trust requirements in social cloud applications.

A Survey on Resource Allocation Strategies in Cloud Computing

Inter-operating grids through Delegated MatchMaking

scholarkusic d, kandasamy n (2006) risk-aware limited lookahead control for dynamic resource provisioning in enterprise computing systems..014mathgoogle scholarxiong p, wang z, malkowski s, wang q, jayasinghe d, pu c (2011) economical and robust provisioning of n-tier cloud workloads: a multi-level control approach. research literature has investigated both centralized and decentralized load balancing mechanisms for providers. this paper aims at supporting these efforts by providing a survey of the state of the art of qos modeling approaches applicable to cloud computing and by describing their initial application to cloud resource management. scholarzou d, zhang w, qiang w, xiang g, yang lt, jin h, & hu k (2013) design and implementation of a trusted monitoring framework for cloud platforms. the proposed solution has the aim to take advantage of a cloud federation to avoid the dependence on a single provider, while still minimizing the amount of used resources to maintain a good qos level for customers. scholarjhawar r, piuri v (2012) fault tolerance management in iaas clouds. while some bias components can be filtered out (for example using the cpu steal metric available on amazon ec2 virtual machines), contention on resources such as cache, memory bandwidth, network, or storage, is harder or even impossible to monitor for the final user. in proceedings of the paper presented at the international conference on image processing, production and computer science, istanbul (turkey). in: proceedings of the 2008 spec international workshop on performance evaluation: metrics, models and benchmarks, sipew ’08 boston, ma, usa, 191–207, darmstadt, germany. on the other hand, while closed-form solutions exist for some classes of queueing systems and queueing networks, the solution of other models, including lqns, rely on numerical methods. the authors formulate an optimization problem faced by a cloud procurement endpoint (a module responsible for provisioning resources from public cloud providers), where heavy workloads are tackled by relying on public clouds. [37] considers the problem of dynamically estimating cpu demands of diverse types of requests using cpu utilization and throughput measurements. [20] uses hidden markov models to capture and predict temporal correlations between workloads of different compute clusters in the cloud. an autoregressive process is used to predict the fluctuating incoming demand. scholarputhal d, sahoo b, mishra s, & swain s (2015) cloud computing features, issues, and challenges: a big picture. scholarchiregi m & navimipour nj (2016a) a new method for trust and reputation evaluation in the cloud environments using the recommendations of opinion leaders’ entities and removing the effect of troll entities. autoregressive models in particular are quite common in applications and they are already exploited in cloud application modeling, e. in: proceedings of 2012 8th international conference on network and service management, and 2012 workshop on systems virtualiztion management, cnsm-svm 2012, 385–392, las vegas, nv, usa. scholarleitner p, hummer w, satzger b, inzinger c, dustdar s (2012) cost-efficient and application sla-aware client side request scheduling in an infrastructure-as-a-service cloud. the performance models are based on queueing-network models abstracted from the system and enhanced by machine learning algorithms to correlate system workload attributes with performance attributes. other works that rely on queueing models to describe cloud resources include [53],[54]. in: proceedings of the 2011 ieee ninth international conference on dependable, autonomic and secure computing, dasc 2011, 598–604, sydney, nsw, australia. scholarpacifici g, segmuller w, spreitzer m, tantawi a: cpu demand for web serving: measurement analysis and dynamic estimation. scholarliu s, huang x, fu h, yang g (2013) understanding data characteristics and access patterns in a cloud storage system. scholarnoor th, sheng qz, maamar z, & zeadally s (2016) managing trust in the cloud: state of the art and research challenges. the proposed algorithm is compared against a greedy heuristic method and it shows significant cost savings (around 20-30%). scholarwang y, chandrasekhar s, singhal m, & ma j (2016) a limited-trust capacity model for mitigating threats of internal malicious services in cloud computing. scholarmao m, humphrey m (2011) auto-scaling to minimize cost and meet application deadlines in cloud workflows. a system model is demonstrated to suitably characterize cloud service provisioning behavior and an exact algorithm is proposed to optimize users’ experience under qos requirements. a comparison against a set of heuristics from the literature and an oracle with perfect knowledge about the future load shows that the proposed algorithm overcomes the heuristic approaches, without penalizing slas and it is able to produce results that are close to the global optimum. a non-linear model for the capacity allocation and load redirection of multiple request classes is proposed and solved by decomposition. this effectively creates a divide between the knowledge that can be made available for an application by its designers and the techniques used to manage it. a queueing network can be described as a collection of queues interacting through request arrivals and departures. extensive validation on generated dataset and real data show the effectiveness of the algorithm. the solution is composed by two parts: first, servers selection in a data-center is performed by using a search based bio-inspired technique; then, data centers are selected within the cloud federation by using a shortest path algorithm, according to the available bandwidth of links connecting the domains. [63] uses an lqn model to predict the performance of the rubis benchmark application, which is then used as the basis of an optimization algorithm that aims at determining the best replication levels and placement of the application components. scholardutta s, gera s, verma a, viswanathan b (2012) smartscale: automatic application scaling in enterprise clouds. scholarjuan d & zheng q (2014) cloud and open bim-based building information interoperability research. this approach proves to be computationally efficient and robust to outliers. scholarwada h, fekete a, zhao l, lee k, liu a (2011) data consistency properties and the trade-offs in commercial cloud storage: the consumers’ perspective. this model predicts service delay, task rejection probability, and steady-state distribution of server pools. other common techniques include wavelet-based methods, regression analysis, filtering, fourier analysis, and kernel-based methods [19]. we survey in section 4 works on decision making for capacity allocation, load balancing, and admission control including research works that provide solutions for the management of a cloud infrastructure (i. scholaralamir p, jafari navimipour n, ramage m, ramage m, & ramage m (2016) trust evaluation between the users of social networks using the quality of service requirements and call log histories. the algorithm developed has the additional benefit to leverage cloud elasticity to allocate and deallocate resources to help services to respect contractual slas. two complementary methods are proposed: an offline deterministic optimization method to be used at design time and an online vm placement, migration and geographical load balancing algorithm for runtime. other common techniques include wavelet-based methods, regression analysis, filtering, fourier analysis, and kernel-based methods [19]..5google scholarcardellini v, casalicchio e, grassi v, mirandola r (2006) a framework for optimal service selection in broker-based architectures with multiple qos classes. in 2010 international conference on computer application and system modeling (iccasm 2010). ability to quantify resource demands is a pre-requisite to parameterize most qos models for enterprise applications. scholarshahzad f (2014) state-of-the-art survey on cloud computing security challenges, approaches and solutions. in: proceedings of the 4th usenix conference on hot topics in cloud ccomputing, hotcloud’12, 4–4, boston, ma, usagoogle scholarschad j, dittrich j, quiané-ruiz j-a: runtime measurements in the cloud: observing, analyzing, and reducing variance. this information is then used as input for cloudsim, which provides qos estimates for a given cloud deployment.

New Roadmap for Elastic Grid Resource Matchmaking

JobSubmissionComparison < Documentation < TWiki

lqns are here useful to handle the complexity of geo-distributed applications that include both transactional and streaming workloads. scholardykstra j & sherman at (2012) acquiring forensic evidence from infrastructure-as-aservice cloud computing: exploring and evaluating tools, trust, and techniques. the authors employ density based clustering to obtain clusters of service times and cpu utilizations, and then use a cluster-wise regression algorithm to estimate the service time. the problem is formulated as a multivariate linear regression problem and accounts for multiple effects such as data aging. the studies considered in the previous section, the load balancer is installed and managed transparently by the cloud provider. [35] proposes the demand estimation with confidence (dec) approach to overcome the problem of multicollinearity in regression methods. we have surveyed current approaches in workload and system modeling and early applications to cloud qos management. modeling these feature usually require an in-depth knowledge of the application behavior. research is needed in this domain to understand the impact of such contention bias on demand estimation. [138] considers an online mechanism for computing resource allocation to vms subject to limited information. more work and validations on paas data are required to mature such techniques. it first uses nonlinear regression to predict the probability for a query to meet its requirement, and then decides whether the query should be admitted to the database system or not. [126] develops a resource manager that uses a combination of horizontal and vertical scaling to optimize both resource usage and the reconfiguration cost. scholarhadji m, zeghlache d (2012) minimum cost maximum flow algorithm for dynamic resource allocation in clouds., a profile-based approach for scalability is described in [109], the authors propose a solution based on the definition of platform-independent profiles, which enable the automation of setup and scaling of application servers in order to achieve a just-in-time scalability of the execution environment, as demonstrated with a case study presented in the paper. forecasting and trend analysis techniques are commonly used to predict web traffic intensity at different timescales. anselmi and casale [120] provides a simple heuristic for user-side load-balancing under connection pooling that is validated against an iaas cloud dataset. [129] also considers the vm consolidation problem by modeling the vm resource demands as a set of correlated random variables. in: international conference on artificial intelligence and statistics, 796–803, sardinia, italy. [18] proposes a model-predictive resource allocation algorithm that auto-scales vms, with the aim of optimizing the utility of the application over a limited prediction horizon.., qos in web services), but it is far less understood in white-box and gray-box modeling. a time based linear clustering algorithm is used to identify different linear clusters for each service demands. scholarfigueiredo j, maciel p, callou g, tavares e, sousa e, silva b (2011) estimating reliability importance and total cost of acquisition for data center power infrastructures. in: proceedings of the 2006 ieee international conference on autonomic computing, icac ’06, 74–83, dublin, ireland. in proceedings of the 2008 acm sigmetrics international conference on measurement and modeling of computer systems. gspns are used to provide fine-grained detail on the inner vm behaviors, such as separation of privileged and non-privileged instructions and successive handling by the vm or the vm monitor. the paper presents a minimum cost maximum flow (mcmf) algorithm and compares it against a modified bin-packing formulation; the mcmf algorithm exhibits very good performance and scalability properties. several works have investigated over the last two decades the problem of estimating, using indirect measurements, the resource demand placed by an application on physical resources, for example cpu requirements. the user perspective, capacity allocation arises in iaas and paas scenarios where the user is in charge with the control of the number of vms or application containers running in the system. he has published more than 100 papers in various journals and conference proceedings. scholarsun d, chang g, sun l, & wang x (2011) surveying and analyzing security, privacy and trust issues in cloud computing environments. here, we survey workload characterization studies and related modeling techniques. in bio-inspired models of network, information, and computing systems, volume 87 of lecture notes of the institute for computer sciences, social informatics and telecommunications engineering. we also review and classify their early application to some decision-making problems arising in cloud qos management. moreover, a closed-form formula for calculating the average response time of a request and a unified framework to manage different levels of slas are provided. many solutions are based on the cloudsim [93] toolkit that allows the user to set up a simulation model that explicitly considers virtualized cloud resources, potentially located in different data centers, as in the case of hybrid deployments.., qos in web services), but it is far less understood in white-box and gray-box modeling. one of the challenges posed by cloud applications is quality-of-service (qos) management, which is the problem of allocating resources to the application to guarantee a service level along dimensions such as performance, availability and reliability. zhu and tung [23] uses a kalman filter to model the interference caused when deploying applications on virtualized resources. centralized approaches, [122] introduces an offline optimization problem for geographical load balancing among data centers, explicitly considering slas and dynamic electricity prices..74google scholaromari t, franks g, woodside m, pan a: efficient performance models for layered server systems with replicated servers and parallel behaviour. theory has the advantage of guaranteeing the stability of the system upon workload changes by modeling the transient behavior and adjusting system configurations within a transitory period [143]. scholarafari navimipour n, rahmani am, navin ah, & hosseinzadeh m (2015) expert cloud: a cloud-based framework to share the knowledge and skills of human resources. scholardutta s, gera s, verma a, viswanathan b (2012) smartscale: automatic application scaling in enterprise clouds. this approach is compared against other techniques and it shows cost savings from 9. scholarnavimipour nj (2015) a formal approach for the specification and verification of a trustworthy human resource discovery mechanism in the expert cloud. however, the diversity of technologies used in cloud systems makes it difficult to analyze their qos and, from the provider perspective, to offer service-level guarantees. [38] proposes an algorithm to estimate the service demands for different system configurations. research work has focused on policies that are either simple to implement, and thus minimize overheads, or that offer some optimality guarantees, typically proven by analytical models. it uses an open multi-class queueing network to support a qos-aware admission control on heterogeneous resources to increase system throughput. scholarzheng t, woodside cm, litoiu m: performance model estimation and tracking using optimal filters. the studies considered in the previous section, the load balancer is installed and managed transparently by the cloud provider. some studies on amazon ec2 have found high-performance contention in cpu-bound jobs [9] and network performance overheads [10]. in computing, electronics and electrical technologies (icceet), 2012 international conference on:877-880.

A Semantic Framework for Resource Discovery Based on Ontology

Resource allocation decision model for dependable and cost

the performance models, we survey queueing systems, queueing networks, and layered queueing networks (lqn). [63] uses an lqn model to predict the performance of the rubis benchmark application, which is then used as the basis of an optimization algorithm that aims at determining the best replication levels and placement of the application components. zhu and tung [23] uses a kalman filter to model the interference caused when deploying applications on virtualized resources. scholargani a, nayeem gm, shiraz m, sookhak m, whaiduzzaman m, & khan s (2014) a review on interworking and mobility techniques for seamless connectivity in mobile cloud computing. mao and humphrey [111] defines an auto-scaling mechanism to guarantee the execution of all jobs within given deadlines. [124] optimizes the allocation and scheduling of vms in federated clouds using a genetic algorithm. scholarcalinescu r, ghezzi c, kwiatkowska mz, mirandola r: self-adaptive software needs quantitative verification at runtime. the performance models, we survey queueing systems, queueing networks, and layered queueing networks (lqn). scholargajbhiye a & shrivastva kmp (2014) cloud computing: need, enabling technology, architecture, advantages and challenges., [105] proposes an adaptive approach for component replication of cloud applications, aiming at finding a cost-effective placement and load balancing. hardware heterogeneity and vm interference are the primary cause for such variability, which is also visible within vms of the same instance class. modeling involves the assessment or prediction of the arrival rates of requests and of the demand for resources (e. the authors also provide a discussion about the pros and cons of lqns identifying a number of key limitations for their practical use in cloud systems. farley b, juels a, varadarajan v, ristenpart t, bowers kd, swift mm (2012) more for your money: exploiting performance heterogeneity in public clouds. approaches are currently being proposed to automate dynamic pricing and cloud resources selection.. cloud computing is an operation model that integrates many technological advancements of the last decade such as virtualization, web services, and sla management for enterprise applications. the basic idea is to predict the value of a specific qos metric and if such value grows above a certain threshold, the admission controller rejects all new sessions favoring the service of requests from already admitted sessions. modeling these feature usually require an in-depth knowledge of the application behavior. [65] proposes a robust performance model architecture focusing on analyzing performance anomalies and localizing the potential source of the discrepancies. [71] investigates the benefits of a warm-standby replication mechanism in eucalyptus cloud computing environments. it has long been recognized the suitability of petri nets for performance and dependability of computer systems. the solution is composed by two parts: first, servers selection in a data-center is performed by using a search based bio-inspired technique; then, data centers are selected within the cloud federation by using a shortest path algorithm, according to the available bandwidth of links connecting the domains. scholarpadala p, shin kg, zhu x, uysal m, wang z, singhal s, merchant a, salem k (2007) adaptive control of virtualized resources in utility computing environments. qos properties have received constant attention well before the advent of cloud computing, performance heterogeneity and resource isolation mechanisms of cloud platforms have significantly complicated qos analysis, prediction, and assurance. yet, the qos modeling literature is extensive, making it difficult to have a comprehensive view of the available techniques and their current applications to cloud computing problems. applications to cloud qos modeling include the use of spns to evaluate the dependability of a cloud infrastructure [68], considering both reliability and availability. a lot of work has been done in the last decade for optimal admission control in web servers and multi-tier applications. scholarpatel n & chauhan s (2015) a survey on load balancing and scheduling in cloud computing. computing has grown in popularity in recent years thanks to technical and economical benefits of the on-demand capacity management model [1]. scholarli x, ma h, zhou f, & gui x (2015) service operator-aware trust scheme for resource matchmaking across multiple clouds. scholarwu x, woodside m (2008) a calibration framework for capturing and calibrating software performance models. scholaraznoli f & and navimipour nj (2017) cloud services recommendation: reviewing the recent advances and suggesting the future research directions. scholarostermann s, plankensteiner k, prodan r, fahringer t (2011) groudsim: an event-based simulation framework for computational grids and clouds. this approach is compared against other techniques and it shows cost savings from 9. a queueing network can be described as a collection of queues interacting through request arrivals and departures. the proposed framework helps to stipulate a realistic sla with customers and supports dynamic load shedding and capacity provisioning by considering a queueing model with multiple priority classes. the authors propose a workload demand prediction algorithm based on trend analysis and pattern recognition. also, this survey paper provides a discussion of differences between considered techniques in terms of integrity, security, availability, reliability, dependability, safety, dynamicity, confidentiality and scalability as well as directions for future research. [32] presents a queueing network model where each queue represents a tier of a web application, which is parameterized by means of a regression-based approximation of the cpu demand of customer transactions. indeed, a trade-off exists between available information, qos model complexity, computational cost of decision-making, and accuracy of predictions. scholarzou d, zhang w, qiang w, xiang g, yang lt, jin h, & hu k (2013) design and implementation of a trusted monitoring framework for cloud platforms. research is needed in this domain to understand the impact of such contention bias on demand estimation. this model predicts service delay, task rejection probability, and steady-state distribution of server pools. availability of resources and admission control is also discussed in [131]. hence, in this paper, the comprehensive and detailed study and survey of the state of the art techniques and mechanisms in this field are provided. scholarrolia j, vetland v: correlating resource demand information with arm data for application services. the authors employ density based clustering to obtain clusters of service times and cpu utilizations, and then use a cluster-wise regression algorithm to estimate the service time. we have surveyed current approaches in workload and system modeling and early applications to cloud qos management. scholartchifilionova v (2011) security and privacy implications of cloud computing-lost in the cloud..21google scholarbacigalupo d, van hemert j, chen x, usmani a, chester a, he l dillenberger d, wills g, gilbert l, jarvis s: managing dynamic enterprise and urgent workloads on clouds using layered queuing and historical performance models. the proposed algorithms consider customer profiles and quality parameters to cope with dynamic workloads and heterogeneous cloud resources. queueing network approach is taken in [66] to provision resources for data-center applications. [49] proposes a joint admission control and capacity allocation algorithm for virtualized iaas systems minimizing the data center energy costs and the penalty incurred for request rejections and sla violations. we consider the resource management mechanisms for applications qos enforcement provided by public clouds, they are quite simplistic if compared to current research proposals.

Amazon GameLift FAQ – Amazon Web Services (AWS)

it uses an open multi-class queueing network to support a qos-aware admission control on heterogeneous resources to increase system throughput. the algorithm reduces resources under- and over-provisioning by minimizing the total cost for a customer during a certain time horizon. scholarel haddad j, manouvrier m, rukoz m: tqos: transactional and qos-aware selection algorithm for automatic web service composition. the proposed solution has the aim to take advantage of a cloud federation to avoid the dependence on a single provider, while still minimizing the amount of used resources to maintain a good qos level for customers. [125] is proposed a framework for vm deployment and reconfiguration optimization, with the aim at increasing profits of iaas providers. [49] proposes a joint admission control and capacity allocation algorithm for virtualized iaas systems minimizing the data center energy costs and the penalty incurred for request rejections and sla violations. in: proceedings of the 2013 13th ieee/acm international symposium on cluster, cloud and grid computing, ccgrid 2013, 327–334, delft, nederlandsgoogle scholarli a, yang x, kandula s, zhang m (2010) cloudcmp: comparing public cloud providers. [138] considers an online mechanism for computing resource allocation to vms subject to limited information. in: proceedings of the 2008 spec international workshop on performance evaluation: metrics, models and benchmarks, sipew ’08 boston, ma, usa, 191–207, darmstadt, germany. the system is represented by a set of inter-related blocks, connected by series, parallel, and k-out-of-n relationships. scholarmenascé d, almeida v, dowdy l: capacity planning and performance modeling: from mainframes to client-server systems. scholarcallou g, maciel p, tutsch d, araujo j (2011) models for dependability and sustainability analysis of data center cooling architectures. two adaptive hybrid controllers, including both reactive and proactive actions, are employed to decide the number of vms for a cloud service to meet the slas. scholarfouladi p & navimipour jn (2017) human resources ranking in a cloud-based knowledge sharing framework using the quality control criteria. many solutions are based on the cloudsim [93] toolkit that allows the user to set up a simulation model that explicitly considers virtualized cloud resources, potentially located in different data centers, as in the case of hybrid deployments. [34] presents an optimization-based inference technique that is formulated as a robust linear regression problem that can be used with both closed and open queueing network performance models.., network bandwidth variance, virtual machine (vm) startup times, start failure probabilities. centralized approaches, [122] introduces an offline optimization problem for geographical load balancing among data centers, explicitly considering slas and dynamic electricity prices..64google scholarbonvin n, papaioannou t, aberer k (2010) an economic approach for scalable and highly-available distributed applications. scholarmilani as & navimipour nj (2016) load balancing mechanisms and techniques in the cloud environments: systematic literature review and future trends. on the other hand, the analysis also shows that turning on the geographical load balancing has a strong impact on quality of the solutions (between 27% and 40%) of the online algorithm. the objectives are the minimization of both resource and penalty costs, as well as minimizing sla violations. a time based linear clustering algorithm is used to identify different linear clusters for each service demands. the authors studied the behavior of both online and offline algorithms by means of a simulation campaign. [33] proposes instead service demand estimation from utilization and end-to-end response times: the problem is formulated as quadratic optimization programs based on queueing formulas; results are in good agreement with experimental data. such mechanism uses biological evolution concepts to manage data application services and to produce optimal composition and load balancing solutions. while queueing systems are widely used to model single resources subject to contention, queueing networks are able to capture the interaction among a number of resources and/or applications components. a system model is demonstrated to suitably characterize cloud service provisioning behavior and an exact algorithm is proposed to optimize users’ experience under qos requirements. scholarhadji m, zeghlache d (2012) minimum cost maximum flow algorithm for dynamic resource allocation in clouds. [70], the authors propose a methodology to evaluate data center power infrastructures considering both reliability and cost. capacity allocation problem in presented in [113] that exploits both horizontal and vertical elasticity.-demand and reserved resources are considered in the model proposed in [107] to define a bio-inspired self-adapting solution for cloud resource provisioning with the aim of minimizing the number of required virtual machines while meeting slas. in: proceedings of the 2011 international conference for high performance computing, networking, storage and analysis, sc ’11, 1–12, seattle, wa, usa. cloud applications are often tiered and queueing networks can capture the interactions between tiers. [69] proposes the use of gspns to evaluate the impact of virtualization mechanisms, such as vm consolidation and live migration, on cloud infrastructure dependability. the load balancer uses the number of outstanding requests and the inter-departure times in each vm to dispatch requests to the vm with the shortest expected response time. we survey in section 4 works on decision making for capacity allocation, load balancing, and admission control including research works that provide solutions for the management of a cloud infrastructure (i. scholarkhan a, yan x, shu t, anerousis n (2012) workload characterization and prediction in the cloud: a multiple time series approach. in: proceedings of the 2013 ieee/ifip 43rd international conference on dependable systems and networks, dsn 2013, 1–6, hong kong, china. [33] proposes instead service demand estimation from utilization and end-to-end response times: the problem is formulated as quadratic optimization programs based on queueing formulas; results are in good agreement with experimental data. each host can run several vms, and has a power model to determine the overall data center power consumption. it has long been recognized the suitability of petri nets for performance and dependability of computer systems. scholardykstra j & sherman at (2012) acquiring forensic evidence from infrastructure-as-aservice cloud computing: exploring and evaluating tools, trust, and techniques. if a component fails, it assumes the logical value true, and the failure propagation can be studied via the tree structure. scholaranselmi, j, ardagna d, & passacantando m (2014) generalized nash equilibria for saas/ paas clouds. scholaraznoli f & and navimipour nj (2017) cloud services recommendation: reviewing the recent advances and suggesting the future research directions. scholarbeloglazov a, buyya r, lee yc, zomaya ay: a taxonomy and survey of energy-efficient data centers and cloud computing systems. the authors use this model to investigate rejection probabilities and help dimensioning of cloud data centers. the problem is formulated as a multivariate linear regression problem and accounts for multiple effects such as data aging. in: proceedings of the 2012 ieee wireless communications and networking conference, wcnc 2012, 3145–3149, paris, france. also, several works have shown how combining the queueing theoretic formulas used by regression methods with the kalman filter can enable continuous demand tracking [41],[42]. [135] considers capacity allocation subject to two pricing models, a pay-as-you-go offering and periodic auctions. the authors reduce costs considering the balance of multi-dimensional resources utilization and building up an optimization method for resource allocation; as far as reconfiguration is concerned, they propose a strategy for vm adjustment based on time-division multiplex and on vm live migration. the user perspective, capacity allocation arises in iaas and paas scenarios where the user is in charge with the control of the number of vms or application containers running in the system.

QoS-aware and Semantic-based Service Coordination for Multi

Multi-Cloud

[65] proposes a robust performance model architecture focusing on analyzing performance anomalies and localizing the potential source of the discrepancies. [90] presents a graph-theoretic model for qos-aware service composition in cloud platforms, explicitly handling network virtualization. policies differ for the decision approach and for the amount of information they use. scholarcalheiros rn, ranjan r, beloglazov a, de rose caf, buyya r: cloudsim: a toolkit for modeling and simulation of cloud computing environments and evaluation of resource provisioning algorithms. the solution accounts for workload burstinesses and delayed instance acquisition. [110] compares several approaches for decision-making, as part of an autonomic framework that allocates resources to a software application. the authors demonstrate that their solution is able to improve the variance and percentiles of response times with respect to a built-in policy of the apache web server. fault trees and markov models are used to evaluate the reliability and availability of fault tolerance mechanisms. scholarzhu q, tung t (2012) a performance interference model for managing consolidated workloads in qos-aware clouds. computing is a new model to enable convenient and on-demand access to the pool of configurable computing resources.., system throughput and utilization of the servers), commonly retrieved from log files, in order to estimate service times. scholarmilani as & navimipour nj (2016) load balancing mechanisms and techniques in the cloud environments: systematic literature review and future trends. in: proceedings of the 2010 ieee symposium on computers and communications, iscc’10, 695–700, riccione, italy. decentralized probabilistic algorithm is also described in [106], which focuses on federated clouds.., when part of the application traffic is redirected from a private to a public data center to cope with a traffic intensity that surpasses the capacity of the private infrastructure), if the public cloud resources are not provided timely, one can decide to drop new incoming request to preserve the qos for users already in the system (or at least part of them, e.., from the cloud provider perspective) and resource management techniques for the infrastructure user (e. queueing theory is used to model different components of the system and data mining and machine learning approaches ensure dynamic adaptation of the model to work under system fluctuations. the authors propose a workload demand prediction algorithm based on trend analysis and pattern recognition. the algorithm evaluates allocation and revenues as the users place requests to the system. computing has grown in popularity in recent years thanks to technical and economical benefits of the on-demand capacity management model [1]. however, finding optimal tradeoff is a difficult decision problem, often exacerbated by the presence of service level agreements (slas) specifying qos targets and economical penalties associated to sla violations [3]..014mathgoogle scholarxiong p, wang z, malkowski s, wang q, jayasinghe d, pu c (2011) economical and robust provisioning of n-tier cloud workloads: a multi-level control approach. then the qos-aware service composition is solved by influence diagrams followed by analytical and simulation experiments. use of this web site signifies your agreement to the terms and conditions.. the aim of this survey is to provide an overview of early research works in the cloud qos modeling space, categorizing contributions according to relevant areas and methods used. bonvin n, papaioannou t, aberer k (2010) an economic approach for scalable and highly-available distributed applications. scholarrolia j, vetland v: correlating resource demand information with arm data for application services. in: proceedings of the 2012 ieee network operations and management symposium, noms 2012, 1287–1294, maui, hi, usa. [38] proposes an algorithm to estimate the service demands for different system configurations. a common workload inference approach involves estimating only the mean demand placed by a given type of requests on the resource [26]-[28]. they present a linear integer program to minimize the resource cost, and evaluate how the solution scales with the different problem parameters. in: proceedings of the 2010 24th ieee international conference on advanced information networking and applications, aina 2010, 446–452, perth, australia. this is complemented with an online algorithm to handle the uncertainty in electricity prices. the authors define nine key features of the workload and use a bayesian classifier to estimate the posterior probability of each feature. scholarkusic d, kephart jo, hanson je, kandasamy n, jiang g: power and performance management of virtualized computing environments via lookahead control. given the lack of control over the system workload and configuration during operation, techniques of this type may not be applicable to production systems for online model calibration. scholargasquet c, witomski p: fourier analysis and applications: filtering, numerical computation, wavelets, volume 30 of texts in applied mathematics. nets, reliability block diagrams (rbd), and fault trees are probably the most widely known and used formalisms for dependability analysis. however, several other classes exist such as stochastic process algebras, stochastic activity networks, stochastic reward nets [44], and models evaluated via probabilistic model checking [45]. moreover, a formulation for the considered problem is presented and its hardness is proven. the experiments are based on a large dataset collected from a google data center with thousands of machines. nevertheless, training sessions tend to extend over several hours [144] and retraining is required for evolving workloads. [125] is proposed a framework for vm deployment and reconfiguration optimization, with the aim at increasing profits of iaas providers. in [127] the capacity allocation problem is solved by means of a dynamic algorithm, since static allocation policies and pricing usually lead to inefficient resource sharing, poor utilization, waste of resources and revenue loss when demands and workloads are time varying. on the other hand, while closed-form solutions exist for some classes of queueing systems and queueing networks, the solution of other models, including lqns, rely on numerical methods.., response time, throughput and resource utilization) predicted by a performance model against measurements collected in a controlled experimental environment. in: proceedings of 2012 international symposium on cloud and services computing, iscos 2012, 25–30, mangalore, india. scholarli z, de souza r, & goh m (2016) supply chain orchestration leveraging on mnc networks and local resources: approach strategies. they use hidden markov models to identify temporal correlations between different clusters and use this information to predict future workload variations. [100] is a simulator for scientific applications deployed on large-scale clouds and grids. scholaralmeida j, almeida v, ardagna d, cunha i, francalanci c, trubian m: joint admission control and resource allocation in virtualized servers. scholarfigueiredo j, maciel p, callou g, tavares e, sousa e, silva b (2011) estimating reliability importance and total cost of acquisition for data center power infrastructures. scholarcallou g, maciel p, tutsch d, araujo j (2011) models for dependability and sustainability analysis of data center cooling architectures. [130] develops an analytical model for resource provisioning, virtual machine deployment, and pool management.

Ontology-based Grid resource management

Quid

work in [139] proposes an admission control protocol to prevent over-utilization of system resources, classifying applications based on resource quality requirements. [71] investigates the benefits of a warm-standby replication mechanism in eucalyptus cloud computing environments..23google scholarkalbasi a, krishnamurthy d, rolia j, richter m (2011) mode: mix driven on-line resource demand estimation. scholarwada h, fekete a, zhao l, lee k, liu a (2011) data consistency properties and the trade-offs in commercial cloud storage: the consumers’ perspective.. in computer engineering, computer architecture, from science and research branch, islamic azad university, tehran, iran in 2014. a comparison with state of the art qos routing algorithms shows that the proposed algorithm is both cost-effective and lightweight. scholarrolia j, vetland v: parameter estimation for performance models of distributed application systems. from the perspective of cloud providers and users, inference techniques provide a means to estimate the workload profile of individual vms running on their infrastructures, taking into account hidden variables due to lack of information. in: proceedings of the 2013 ieee 5th international conference on cloud computing technology and science, volume 1 of cloudcom 2013, 1–8, bristol, united kingdom. scholargani a, nayeem gm, shiraz m, sookhak m, whaiduzzaman m, & khan s (2014) a review on interworking and mobility techniques for seamless connectivity in mobile cloud computing. an extensible meta-model and a class library with an initial set of five models are developed. in cloud computing, fault trees have been used to evaluate dependencies of cloud services and their effect on application reliability [73]. indeed, such mechanisms are mainly reactive and are triggered by thresholds violations (related to response times, as in google app engine, or cpu utilization or other low level infrastructure metrics, as in amazon ec2. in: proceedings of the 2011 ieee ninth international conference on dependable, autonomic and secure computing, dasc 2011, 598–604, sydney, nsw, australia. ps scheduling is assumed at the resources to model cpu sharing. availability of resources and admission control is also discussed in [131]. techniques to determine optimized decisions range from simple heuristics to nonlinear programming and meta-heuristics. the authors propose also a faster near-optimal algorithm, proven to asymptotically approach the optimal solution, and show a significantly lower complexity with respect to the optimal method. also, [56] uses a queueing network to represent a multi-tier application deployed in a cloud platform, and to derive an sla-aware resource allocation policy. in computational intelligence and networks (cine), 2015 international conference on:116-123. in: proceedings of the 2012 ieee wireless communications and networking conference, wcnc 2012, 3145–3149, paris, france. scholarkalbasi a, krishnamurthy d, rolia j, dawson s: dec: service demand estimation with confidence. [21] defines a bayesian algorithm for long-term workload prediction and pattern analysis, validating results on data from a google data center. a few characterization studies specific to public and private paas hosting solutions also appeared in the literature [11],[12], together with comparisons of cloud database and storage services [13]-[16]. scholarwu x, woodside m (2008) a calibration framework for capturing and calibrating software performance models..066google scholarfranks g, al-omari t, woodside cm, das o, derisavi s: enhanced modeling and solution of layered queueing networks. scholargmach d, rolia j, cherkasova l, kemper a (2007) workload analysis and demand prediction of enterprise data center applications. in: proceedings of 2012 8th international conference on network and service management, and 2012 workshop on systems virtualiztion management, cnsm-svm 2012, 385–392, las vegas, nv, usa. cloud computing centralized distributed monitoring and tracking matin chiregi received her b. finally, the presented approach is endorsed against fixed and adaptive control schemes by means of a campaign of experiments. surprisingly, we have found a limited amount of work specific to workload analysis and inference techniques in the cloud. scholarliu s, huang x, fu h, yang g (2013) understanding data characteristics and access patterns in a cloud storage system. a refinement process is conducted between clustering and regression to get accurate clustering results by removing outliers and merging the clusters that fit the same model. moreover, a formulation for the considered problem is presented and its hardness is proven. qos denotes the levels of performance, reliability, and availability offered by an application and by the platform or infrastructure that hosts ita. the basic idea is to predict the value of a specific qos metric and if such value grows above a certain threshold, the admission controller rejects all new sessions favoring the service of requests from already admitted sessions. [126] develops a resource manager that uses a combination of horizontal and vertical scaling to optimize both resource usage and the reconfiguration cost. cloudanalyst [94] is an extension of cloudsim that allows the modeling of geographically-distributed workloads served by applications deployed on a number of virtualized data centers. research literature has investigated both centralized and decentralized load balancing mechanisms for providers. the authors propose also a faster near-optimal algorithm, proven to asymptotically approach the optimal solution, and show a significantly lower complexity with respect to the optimal method. they present a linear integer program to minimize the resource cost, and evaluate how the solution scales with the different problem parameters. this may be helpful, for instance, to jointly tackle capacity allocation and load balancing. scholarsowmya k, sundarraj rp (2012) strategic bidding for cloud resources under dynamic pricing schemes. scholaragostinho l, feliciano g, olivi l, cardozo e, guimaraes e (2011) a bio-inspired approach to provisioning of virtual resources in federated clouds. scholarkalbasi a, krishnamurthy d, rolia j, dawson s: dec: service demand estimation with confidence. furthermore, we defined trust characteristics such as integrity, security, availability, reliability, dependability, safety, dynamicity, confidentiality and scalability, and we discuss the trust applications including monitoring and tracking. this information is then used as input for cloudsim, which provides qos estimates for a given cloud deployment. scholarko rk, jagadpramana p, mowbray m, pearson s, kirchberg m, liang q (2011) trustcloud: a framework for accountability and trust in cloud computing. [141] proposes an online load balancing policy that considers the inherent vm heterogeneity found in cloud resources. [95] builds on top of cloudsim by adding an emulation step leveraging the automated emulation framework (aef) [96]. in: proceedings of the 7th international conference on network and services management, 1–9. an analytical model, based on queueing theory, is presented to describe the relation between the number of replicas and the service level, e. in: proceedings of the 6th international symposium on software engineering for adaptive and self-managing systems, seams ’11, 218–227, honolulu, hi, usa. in [127] the capacity allocation problem is solved by means of a dynamic algorithm, since static allocation policies and pricing usually lead to inefficient resource sharing, poor utilization, waste of resources and revenue loss when demands and workloads are time varying.

Dating a girl with cervical cancer

Managing elasticity across multiple cloud providers

the solution accounts for workload burstinesses and delayed instance acquisition. in particular, the m/g/1 ps queue is a common abstraction used to model a cpu and it has been adopted in many cloud studies [47],[48], thanks to its simplicity and the suitability to apply the model to multi-class workloads. other works that rely on queueing models to describe cloud resources include [53],[54]. scholarnoor th, sheng qz, maamar z, & zeadally s (2016) managing trust in the cloud: state of the art and research challenges. in: proceedings of the 2011 ieee 13th international conference on high performance computing and communications, hpcc 2011, 784–789, bamff, canada. scholarardagna d, casolari s, colajanni m, panicucci b: dual time-scale distributed capacity allocation and load redirect algorithms for cloud systems. the experiments are based on a large dataset collected from a google data center with thousands of machines. petri nets have been extended to consider stochastic transitions, in stochastic petri nets (spns) and generalized spns (gspns). this is complemented with an online algorithm to handle the uncertainty in electricity prices., a profile-based approach for scalability is described in [109], the authors propose a solution based on the definition of platform-independent profiles, which enable the automation of setup and scaling of application servers in order to achieve a just-in-time scalability of the execution environment, as demonstrated with a case study presented in the paper. also, this survey paper provides a discussion of differences between considered techniques in terms of integrity, security, availability, reliability, dependability, safety, dynamicity, confidentiality and scalability as well as directions for future research. scholarardagna d, casolari s, colajanni m, panicucci b: dual time-scale distributed capacity allocation and load redirect algorithms for cloud systems. a few characterization studies specific to public and private paas hosting solutions also appeared in the literature [11],[12], together with comparisons of cloud database and storage services [13]-[16]. scholarvan den, bossche r, vanmechelen k, broeckhove j (2010) cost-optimal scheduling in hybrid iaas clouds for deadline constrained workloads.. in computer engineering, computer architecture, from science and research branch, islamic azad university, tehran, iran in 2014. scholarranjan r, zhao l, wu x, liu a, quiroz a, parashar m: peer-to-peer cloud provisioning: service discovery and load-balancing. [118] introduces a client-side admission control method to schedule requests among vms, looking at minimizing the cost of application, sla violations and iaas resources. scholaralmeida j, almeida v, ardagna d, cunha i, francalanci c, trubian m: joint admission control and resource allocation in virtualized servers. his research interests include cloud computing, social networks, fault-tolerance software, computational intelligence, evolutionary computing, and network on chip. therefore, one of the most important challenges in this environment is to evaluate the trust value to enable users for selecting the trustworthy resources, however, to the best of our knowledge, the comprehensive and detailed review of the most important techniques in this field is very rare. the main result is that the presented approach is able to provide tight guarantees on the optimality gap and experimental results show that it is at the same time accurate and fast. there is a growing interest towards understanding better cloud spot markets, where bidding strategies are developed for procuring computing resources. scholarsadashiv n & kumar sd (2011) cluster, grid and cloud computing: a detailed comparison..284mathscinetgoogle scholarzaman s, grosu d (2012) an online mechanism for dynamic vm provisioning and allocation in clouds. ps scheduling is assumed at the resources to model cpu sharing.., cpu requirements) placed by applications on an infrastructure or platform, and the qos observed in response to such workloads. [135] considers capacity allocation subject to two pricing models, a pay-as-you-go offering and periodic auctions. moreover, it improves the placement of application instances by putting idle machines into standby mode and reducing the number of running instances in condition of light load. autoregressive models in particular are quite common in applications and they are already exploited in cloud application modeling, e. in proceedings of the 1st international workshop on software and performance. this approach proves to be computationally efficient and robust to outliers. however, several other classes exist such as stochastic process algebras, stochastic activity networks, stochastic reward nets [44], and models evaluated via probabilistic model checking [45]. several works instead adopt a description that includes standard deviations [76],[82],[83] or finite ranges of variability for the execution times [84],[85]. more work and validations on paas data are required to mature such techniques. yet, the qos modeling literature is extensive, making it difficult to have a comprehensive view of the available techniques and their current applications to cloud computing problems. [141] proposes an online load balancing policy that considers the inherent vm heterogeneity found in cloud resources. [43] studies the relations between workload and resource consumption for cloud web applications. the work uses a probabilistic approach to find an optimized allocation of services on virtualized physical resources. challenges a threat to workload inference on iaas clouds is posed by resource contention by other users, which can systematically result in biased readings of performance metrics. scholarostermann s, plankensteiner k, prodan r, fahringer t (2011) groudsim: an event-based simulation framework for computational grids and clouds. two adaptive hybrid controllers, including both reactive and proactive actions, are employed to decide the number of vms for a cloud service to meet the slas. we survey in section 3 formalisms and tools employed for these analyses and their current applications to assess the performance of cloud systems. the authors define nine key features of the workload and use a bayesian classifier to estimate the posterior probability of each feature. the consolidation algorithm is tested and shown to be highly competitive. the proposed approach is shown to achieve high accuracy for predicting workload and resource usages. scholarkhazaei h, misic j, misic v, rashwand s: analysis of a pool management scheme for cloud computing centers. this is prompting several researchers to investigate automated qos management methods that can leverage the high programmability of hardware and software resources in the cloud [4]. approaches are currently being proposed to automate dynamic pricing and cloud resources selection. common analytical formulas involve queues with exponential service and arrival times, with a single server (m/m/1) or with k servers (m/m/k), and queues with generally-distributed service times (m/g/1). these include, among others, difficulties in modeling caching, lack of methods to compute percentiles of response times, tradeoff between accuracy and speed. scholarzhu q, tung t (2012) a performance interference model for managing consolidated workloads in qos-aware clouds. indeed, such mechanisms are mainly reactive and are triggered by thresholds violations (related to response times, as in google app engine, or cpu utilization or other low level infrastructure metrics, as in amazon ec2. then, the problem to be addressed is to determine the minimum number of vms or containers needed to fulfill the target qos, pursuing the best trade-off between cost and performance. compared to ordinary queueing networks, lqns provide the ability to describe dependencies arising in a complex workflow of requests and the layering among hardware and software resources that process them.

Home Sitemap