PERFORMANCE EVALUATION AND MODELLING OF SAAS WEB SERVICES IN THE CLOUD

Doctoral thesis (2020)

This thesis studies the problem of performance evaluation and assurance of communications and services quality in cloud computing. The cloud computing paradigm has significantly changed the way of doing ... [more ▼]

This thesis studies the problem of performance evaluation and assurance of communications and services quality in cloud computing. The cloud computing paradigm has significantly changed the way of doing business. With cloud computing, companies and end-users can access the vast majority of services online through a virtualized environment. The main three services typically consumed by cloud users are Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS) and Software-as-a-Service (SaaS). Cloud Services Providers (CSPs) deliver cloud services to cloud customers on a pay-per-use model while the quality of the provided services is defined using Service Level Agreements (SLAs). Unfortunately, there is no standard mechanism which exists to verify and assure that delivered services satisfy the signed SLA agreement in an automatic way, which impedes the possibility to measure accurately the Quality of Service (QoS). In this context, this thesis aims at offering an automatic framework to evaluate the QoS and SLA compliance of Web Services (WSs) offered across several CSPs. Yet unlike other approaches, the framework aims at quantifying in a fair and by stealth way the performance and scalability of the delivered WS. By stealthiness, it refers to the capacity of evaluating a given Cloud service through multiple workload patterns that makes them indistinguishable from regular user traffic from the provider point of view. This thesis work is motivated by recent scandals in the automotive sector, which demonstrate the capacity of solution providers to adapt to the behavior of their product when submitted to an evaluation campaign to improve the performance results. The framework defines a set of Common performance metrics handled by a set of agents within customized clients for measuring the behavior of cloud applications on top of a given CSP. Once modeled accurately, the agent behavior can be dynamically adapted to hide the true nature of the framework client to the CSP. In particular, the following contributions are proposed: • A new framework of performance metrics for communication systems of cloud computing SaaS. The proposed framework evaluates and classifies in a fair and stealth way the performance and scalability of the delivered WS across multiple CSPs. • Analysis of the performance metrics for the cloud SaaS Web Service (WS) by analyzing all the possible metrics which could be used to evaluate and monitor the behavior of the cloud applications. • Benchmarking the cloud SaaS applications and web services by using referenced benchmarking tools and frameworks. • Modeling the SaaS WS by providing a set of Gaussian models. These models can be used to help the other researchers to generate data representing the CSP’s behavior under a high load and under the normal usage in just a couple of minutes without any experiments. • A novel optimization model to obfuscate the testing from the CSP and to achieve stealthiness. The optimization process relies on meta-heuristic and machine learning algorithms, such as Genetic Algorithm and Gaussian Process Regression accordingly. • A virtual QoS aggregator and SLA checker which takes care of evaluating the QoS and SLA compliance of the WS offered across the considered CSPs. • Ranking multiple CSPs based on multi-criteria decision analysis. [less ▲]

Security, reliability and regulation compliance in Ultrascale Computing System

in Zomaya, A. Y.; Carretero, J.; Jeannot, E. (Eds.) Ultrascale Computing Systems (2019)

Ultrascale Computing Systems (UCSs) are envisioned as large-scale complex systems joining parallel and distributed computing systems that will be two to three orders of magnitude larger than today’s ... [more ▼]

Ultrascale Computing Systems (UCSs) are envisioned as large-scale complex systems joining parallel and distributed computing systems that will be two to three orders of magnitude larger than today’s systems (considering the number of Central Process Unit (CPU) cores). It is very challenging to find sustainable solutions for UCSs due to their scale and a wide range of possible applications and involved technologies. For example, we need to deal with heterogeneity and cross fertilization among HPC, large-scale distributed systems, and big data management. One of the challenges regarding sustainable UCSs is resilience. Another one, which attracted less interest in the literature but becomes more and more crucial with the expected convergence with the Cloud computing paradigm, is the notion of regulation in such system to assess the Quality of Service (QoS) and Service Level Agreement (SLA) proposed for the use of these platforms. This chapter covers both aspects through the reproduction of two articles: [1] and [2]. [less ▲]

PRESENCE: Performance Metrics Models for Cloud SaaS Web Services

in Proc. of the 11th IEEE Intl. Conf. on Cloud Computing (CLOUD 2018) (2018, July)

On Verifying and Assuring the Cloud SLA by Evaluating the Performance of SaaS Web Services Across Multi-cloud Providers

in 48th Annual IEEE/IFIP Intl. Conf. on Dependable Systems and Networks Workshops (DNS'18) (2018, June)

Self-Regulated Multi-criteria Decision Analysis: An Autonomous Brokerage-Based Approach for Service Provider Ranking in the Cloud

in 9th IEEE International Conference on Cloud Computing Technology and Science (CloudCom 2017), December 11-14, Hong Kong China. (2017)

The use of multi-criteria decision analysis (MCDA) by online broker to rank different service providers in the Cloud is based upon criteria provided by a customer. However, such ranking is prone to bias ... [more ▼]

The use of multi-criteria decision analysis (MCDA) by online broker to rank different service providers in the Cloud is based upon criteria provided by a customer. However, such ranking is prone to bias if the customer has insufficient domain knowledge. He/she may exclude relevant or include irrelevant criterion termed as ’misspecification of criterion’. This causes structural uncertainty within the MCDA leading to selection of suboptimal service provider by online broker. To cater such issue, we propose a self-regulated MCDA, which uses notion of factor analysis from the field of statistics. Two QoS based datasets were used for evaluation of proposed model. The prior dataset i.e., feedback from customers, was compiled using leading review websites such as Cloud Hosting Reviews, Best Cloud Computing Providers, and Cloud Storage Reviews and Ratings. The later dataset i.e., feedback from servers, was generated from Cloud brokerage architecture that was emulated using high performance computing (HPC) cluster at University of Luxembourg (HPC @ Uni.lu). The results show better performance of proposed model as compared to its counterparts in the field. The beneficiary of the research would be enterprises that view insufficient domain knowledge as a limiting factor for acquisition of Cloud services. [less ▲]

Using Virtual Desktop Infrastructure to Improve Power Efficiency in Grinfy System

in IEEE 8th International Conference on Cloud Computing Technology and Science(CloudCom), Luxembourg 2016 (2016, December 13)

Saving power becomes one of the main objectives in information technology industry and research. Companies consume a lot of money in the shape of power consuming. Virtual Desktop Infrastructure (VDI) is a ... [more ▼]

Saving power becomes one of the main objectives in information technology industry and research. Companies consume a lot of money in the shape of power consuming. Virtual Desktop Infrastructure (VDI) is a new shape of delivering operating systems remotely. Operating systems are executing in a cloud data center. Users desktops and applications can be accessed by using thin client devices. Thin client device is consisting of screen attached with small CPU. VDI has benefits in terms of cost reduction and energy saving. In this paper, we increase the power saved by Grinfy system. Without VDI, Grinfy can save at least 30% of energy consumption to its users companies. By integrating VDI in computing systems and using Grinfy, the power efficiency and saving can be improved and save more than 30%. The improving and increasing of energy saving features of VDI are also illustrated by experiment and will be integrated to Grinfy system to increase percentage of energy saved. [less ▲]

On Service Level Agreement Assurance in Cloud Computing Data Centers

in IEEE 9th International Conference on Cloud Computing (CLOUD), San Francisco, USA (June 2016) (2016, June 27)

Cloud computing uses internet data centers to host applications and data storage. Cloud computing resources and services are offered to customers on pay-per-use model while the quality of the offered ... [more ▼]

Cloud computing uses internet data centers to host applications and data storage. Cloud computing resources and services are offered to customers on pay-per-use model while the quality of the offered resources and services are defined using service level agreements also known as SLAs. Unfortunately, there is no standard mechanism to verify and assure that services delivered by the cloud provider satisfy the SLA agreement in an automatic way. To fill this gap we propose a framework for SLA assurance, which can be used by both cloud providers and cloud users. The proposed framework assesses performance of cloud applications with and without introducing system and component failures and then helps to resolve or mitigate failures to assure the required quality of cloud applications. The evaluation results obtained through simulations and using testbed experiments demonstrate good agreement with the design objectives. [less ▲]

Service Level Agreement Assurance in Cloud Computing Data Centers

Bachelor/master dissertation (2015)

SaaS is providing cloud applications like all the normal classes of applications of normal computing like the web applications, file applications, email applications, real-time applications, highly ... [more ▼]

SaaS is providing cloud applications like all the normal classes of applications of normal computing like the web applications, file applications, email applications, real-time applications, highly interactive applications, massive data analysis applications, high performance computing applications and mobile cloud applications. Cloud computing uses the internet data centers to host the applications and data storage and also the processing power in the addition to the virtualization. Clouds are a huge stack of easily and usefully virtualized services and resources (Like software, platform and hardware). These resources are used by people all over the world and dynamically configured to accommodate more and more load and scale to a huge number of users. This stack of resources and services is delivered to the customers by a pay-per-use model which verifies that the services provided by the cloud providers are provided by means of service level agreements. In this master thesis, we present an assurance to the service level agreement between the cloud users and the cloud services providers by trying to assess and evaluate the cloud computing data centers services and applications provided to the cloud users. Also we introduce mathematical test models for the web, file, real time and distributed applications. We can use these test models to see the behavior of the applications over a long time. In this master thesis, we classify the cloud applications into 8 different classes of applications and then identify the failures and problems which affect these cloud applications. We classify also all failures types which may occur in cloud data centers like network failures, physical server failures, VM unavailability, individual failures racks and individual component failures. We got the important metrics for each class of applications and developed real scenarios for four classes. Then, we introduce different failures to these scenarios to see what is the impact in the most important metrics of applications. Then, we simulate these scenarios in Network Simulator 2. Finally, we solved and mitigated the failures in simulation by using two failure mitigation techniques. The first one is the virtual machine migration and redundancy. The second technique is the forward error correction. Experimental results acquired over the course of this master's thesis give an assessment and verification of the cloud applications services provided by the cloud provider data centers. And also verify that the cloud data center provider gives their users the best services performance and high QoS, we use this verification to assure SLA response times metric and a good quality and performance of services as mentioned inside the SLA document. This verification will be given to the cloud customers. [less ▲]