Por favor, use este identificador para citar o enlazar este ítem:
https://hdl.handle.net/20.500.12104/83340Registro completo de metadatos
| Campo DC | Valor | Lengua/Idioma |
|---|---|---|
| dc.contributor.advisor | González Pico, Mario Ángel Siller | |
| dc.contributor.advisor | González Castolo, Juan Carlos | |
| dc.contributor.advisor | Meda Campaña, María Elena | |
| dc.contributor.author | Ruíz Díaz, Carlos Alberto | |
| dc.date.accessioned | 2021-10-02T20:27:06Z | - |
| dc.date.available | 2021-10-02T20:27:06Z | - |
| dc.date.issued | 2018-01-10 | |
| dc.identifier.uri | https://wdg.biblio.udg.mx | |
| dc.identifier.uri | https://hdl.handle.net/20.500.12104/83340 | - |
| dc.description.tableofcontents | Contents Summary iii Acknowledgements v List of Figures xi List of Tables xiii 1. Introduction and Motivation 1 1.1. Introduction................................... 1 1.2. Motivation ................................... 3 1.2.1. Definitionofthecontextdomain ................... 3 1.2.2. Definitionoftheresearchproblem .................. 4 1.2.3. Relevanceoftheresearchproblem .................. 6 1.2.4. Hypothesisandresearchquestions .................. 7 1.2.5. Researchgoal.............................. 9 1.2.6. Researchworkscopeandassumptions . . . . . . . . . . . . . . . . 11 1.2.7. Structureofthethesis......................... 12 2. Theoretical Context 13 2.1. SoftwareProductLines ............................ 13 2.2. ModelDrivenEngineering........................... 14 2.3. Cloudcomputing................................ 15 2.4. Virtualization.................................. 17 2.5. Self-adaptivesystems.............................. 18 2.5.1. Cloud computing and self-adaptive systems . . . . . . . . . . . . . 18 2.6. PerformancePrediction ............................ 19 2.6.1. Platformdomainsubcategory..................... 20 2.6.2. Predictionmethodcategory...................... 20 2.6.3. RLSestimation............................. 21 2.7. Relatedwork .................................. 23 2.7.1. SystematizationthroughSPLsandMDE . . . . . . . . . . . . . . 23 2.7.2. ResourceefficiencythroughVMadaptation . . . . . . . . . . . . . 25 3. XIPE Framework 29 3.1. Architecturaldesign .............................. 29 3.1.1. Creation and deployment of IaaS cloud configurations . . . . . . . 30 A SPL-based Approach for the Configuration and Adaptation of IaaS Deployments 3.2. Userinterface(UI) ............................... 31 3.3. Corecomponent(C-SPL) ........................... 32 3.3.1. SPLmanager.............................. 32 3.3.2. Modelmanager............................. 32 3.3.3. Adaptationmanager.......................... 34 3.4. Communicationcomponent(CC)....................... 37 3.5. Implementation................................. 38 3.5.1. Userinterface(UI)........................... 39 3.5.2. Corecomponent(C-SPL) ....................... 39 3.5.3. Communicationcomponent(CC)................... 45 3.5.4. Testcasescenario ........................... 46 4. XIPE Framework evaluation 49 4.1. Qualitative Evaluation of the SPL-based configuration Approach . . . . . 49 4.2. Quantitative evaluation of the Adaptation Capabilities of the Framework 49 4.2.1. Environmentsetup........................... 50 4.2.2. EvaluationofBaseCase........................ 51 4.2.3. EvaluationoftheXIPEFramework ................. 52 4.2.4. EvaluationoftheOpenStackSolution . . . . . . . . . . . . . . . . 54 4.2.5. Apdexsatisfactionindex........................ 55 4.2.6. StatisticalSignificancetests...................... 57 4.3. Discussion.................................... 58 5. Conclusions and future work 59 5.1. Thesissummary ................................ 59 5.1.1. Chapter1. IntroductionandMotivation . . . . . . . . . . . . . . . 59 5.1.2. Chapter2.TheoreticalContext.................... 59 5.1.3. Chapter3.XIPEFramework ..................... 59 5.1.4. Chapter4. XIPEFrameworkevaluation . . . . . . . . . . . . . . . 60 5.2. Contributions.................................. 60 5.2.1. Researchquestions........................... 60 5.2.2. Contributions.............................. 62 5.3. Futurework................................... 63 5.3.1. Supportforcomplexcloudconfigurations . . . . . . . . . . . . . . 63 5.3.2. Integrationofinter-cloudsdeployments . . . . . . . . . . . . . . . 64 5.3.3. Extension of the proposed framework to allow for horizontal and verticalscaling ............................. 64 5.3.4. Adaptationofmultipleresources ................... 64 5.3.5. Supportformultipleadaptationpolicies . . . . . . . . . . . . . . . 64 5.4. ConcludingRemarks.............................. 65 Appendix A. Proposed solution comparison 67 A.1. Proposed solution comparison with previous approaches . . . . . . . . . . 67 viii Ru ́ız Carlos, PhD Thesis, University of Guadalajara, CUCEA, 2018 Appendix B. Cloud test bed 75 B.1.Hardwareconfiguration ............................ 75 B.1.1. Clustercontrollernode......................... 75 B.1.2. Clustercomputenode ......................... 75 Appendix C. VM Base case 79 C.1.VMBasecaseevaluation............................ 79 C.1.1. VMBasecaseprofiling ........................ 79 C.1.2. Profilingunderzeroworkloadconditions. . . . . . . . . . . . . . . 80 C.1.3. Profiling under synthetic workload conditions . . . . . . . . . . . . 80 C.1.4.Results ................................. 81 C.2. DeterminationofVMbasecasesaturationpoint . . . . . . . . . . . . . . 83 Appendix D. Prediction Technique 85 D.1.RLSalgorithm ................................. 85 D.1.1. Recursive Least Square algorithm (RLS algorithm) . . . . . . . . . 85 D.2.RLSparameters ................................ 87 D.3. MemoryasindicatortodescribeVMbehaviour . . . . . . . . . . . . . . D.3.1. Profiling of Memory-based VM configurations . . . . . . . . . . . Appendix E. Proposed solution evaluation . 89 . 90 95 E.1. Solutionbehaviourundersyntheticworkload . . . . . . . . . . . . . . . E.1.1. Performance comparison of the proposed solution to other VM . 95 configurations.............................. 95 Appendix F. OpenStack evaluation 99 F.1.OpenStackauto-scalingconfigurations.................... 99 F.1.1. OpenStackconfiguration........................ 99 Appendix G. Publications 101 G.1. An RLS Memory-based Mechanism for the Automatic Adaptation of VMs onCloudEnvironments ............................101 G.2. Towards a Software Product Line-based approach to adapt IaaS cloud configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 Bibliography 103 A SPL-based Approach for the Configuration and Adaptation of IaaS Deployments x Ru ́ız Carlos, PhD Thesis, University of Guadalajara, CUCEA, 2018 List of Figures 2.1. GraphicalrepresentationoftheRLSalgorithm . . . . . . . . . . . . . . . 23 3.1. XIPEframework ................................ 29 3.2. XIPEpredictionmechanism.......................... 35 3.3. Adaptationprocesstimedautomaton..................... 38 3.4. XIPEFeaturetreemetamodel......................... 40 4.1. XIPESPLmodelimplementation....................... 50 4.2. XIPEbehaviourunderrealworkloadconditions. . . . . . . . . . . . . . . 54 4.3. OpenStacksolutionresponsetime....................... 55 D.1.Predictioninterval5seconds ......................... 85 D.2.Predictioninterval20seconds......................... 86 D.3.Predictioninterval60seconds......................... 86 E.1.XIPEresponsetimecomparisonresults ................... 96 E.2.XIPEthroughputcomparisonresults..................... 96 A SPL-based Approach for the Configuration and Adaptation of IaaS Deployments xii Ru ́ız Carlos, PhD Thesis, University of Guadalajara, CUCEA, 2018 List of Tables 1.1. Researchvariables ............................... 10 2.1. RLSalgorithmparameters........................... 22 3.1. Model-to-Modelmappingrules ........................ 43 3.2. XIPEresourcedatasample .......................... 44 3.3. XIPEresponsetimedatasample ....................... 45 4.1. Proposedsolutionperformance ........................ 53 4.2. OpenStacksolutionperformance ....................... 56 4.3. ApdexfortheconfigurationOp6020ofOpenstack . . . . . . . . . . . . . 56 4.4. ApdexfortheconfigurationOp7020ofOpenstack . . . . . . . . . . . . . 56 4.5. ApdexfortheconfigurationOp8020ofOpenstack . . . . . . . . . . . . . 56 4.6. Apdexfortheproposedsolution ....................... 57 4.7. One-sample t-test for OpenStack and proposed solution . . . . . . . . . . 57 4.8. Two-sample t-test for OpenStack and proposed solution . . . . . . . . . . 58 A.1. Automation and systematization through the use of SPL and MDE tech- niquesPartA.................................. 68 A.2. Automation and systematization through the use of SPL and MDE tech- niquesPartB.................................. 69 A.3. Automation and systematization through the use of SPL and MDE tech- niquesPartC.................................. 70 A.4. Adaptation and resource efficiency (Vertical scaling) Part A . . . . . . . . 71 A.5. Adaptation and resource efficiency (Vertical scaling) Part B . . . . . . . . 72 A.6. Adaptation and resource efficiency (Vertical scaling) Part C . . . . . . . . 73 B.1. Hardware characteristics of the cluster controller node . . . . . . . . . . . 76 B.2. Hardware characteristics of the cluster compute node . . . . . . . . . . . . 77 C.1.Requestrateandworkloadmixture...................... 80 C.2. VMbehaviourunderzeroworkloadconditions . . . . . . . . . . . . . . . 81 C.3.Maximumrequestrate2840req/sec .................... 82 C.4.Maximumrequestrate2850req/sec .................... 82 C.5.Maximumrequestrate2860req/sec .................... 82 C.6. VM Base case performance under distinct request rates . . . . . . . . . . 84 D.1. Predicted values and error (5 sec prediction interval) . . . . . . . . . . . 87 D.2.RLS-basedpredictorparameters ....................... 88 D.3.Adaptationparameters............................. 88 D.4.VMbehaviour(1vCPUand1GBMemory) ................. 91 D.5.VMbehaviour(1vCPUand1.5GBMemory) ................ 92 D.6.Correlationcoefficients............................. 92 D.7.Correlationcoefficients............................. 92 E.1. Proposed solution performance under synthetic workload . . . . . . . . . . 96 F.1. OpenStack auto-scaling configurations . . . . . . . . . . . . . . . . . . . . 100 | |
| dc.format | application/PDF | |
| dc.language.iso | spa | |
| dc.publisher | Biblioteca Digital wdg.biblio | |
| dc.publisher | Universidad de Guadalajara | |
| dc.rights.uri | https://www.riudg.udg.mx/info/politicas.jsp | |
| dc.subject | A Spl | |
| dc.title | A SPL-based Approach for the Configuration and Adaptation of laaS Deployments. | |
| dc.type | Tesis de Doctorado | |
| dc.rights.holder | Universidad de Guadalajara | |
| dc.rights.holder | Ruíz Díaz, Carlos Alberto | |
| dc.coverage | ZAPOPAN, JALISCO | |
| dc.type.conacyt | doctoralThesis | |
| dc.degree.name | DOCTORADO EN TECNOLOGIAS DE INFORMACION | |
| dc.degree.department | CUCEA | |
| dc.degree.grantor | Universidad de Guadalajara | |
| dc.rights.access | openAccess | |
| dc.degree.creator | DOCTOR EN TECNOLOGIAS DE INFORMACION | |
| dc.contributor.director | Durán Limón, Héctor Alejandro | |
| Aparece en las colecciones: | CUCEA | |
Ficheros en este ítem:
| Fichero | Tamaño | Formato | |
|---|---|---|---|
| DCUCEA10071FT.pdf | 1.63 MB | Adobe PDF | Visualizar/Abrir |
Los ítems de RIUdeG están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.