Performance Evaluation Of Cloud Database And Traditional Database In .

1y ago
50 Views
3 Downloads
875.89 KB
53 Pages
Last View : 2d ago
Last Download : 2m ago
Upload by : Javier Atchley
Transcription

Master ThesisElectrical EngineeringDecember 2012Performance Evaluation of Cloud Database andTraditional Database in terms of Response Timewhile Retrieving the DataKaushik DonkenaSubbarayudu GannamaniSchool of ComputingBlekinge Institute of Technology371 79 KarlskronaSweden

This thesis is submitted to the School of Computing at Blekinge Institute of Technology inpartial fulfillment of the requirements for the degree of Master of Science in ElectricalEngineering with emphasis on Electrical Engineering. The thesis is equivalent to 20 weeks offull time studies.Contact Information:Authors:Kaushik DonkenaE-mail: kaushik2d@gmail.comSubbarayudu GannamaniE-mail: rayudu218@gmail.comUniversity advisor:Prof. Lars LundbergSchool of ComputingE-mail: lars.lundberg@bth.seSchool of ComputingBlekinge Institute of Technology371 79 KarlskronaSwedenInternetPhoneFax: www.bth.se/com: 46 455 38 50 00: 46 455 38 50 57ii

ABSTRACTContext: There has been an exponential growth in the size of the databases in the recenttimes and the same amount of growth is expected in the future. There has been a firm drop inthe storage cost followed by a rapid increase in the storage capacity. The entry of Cloud inthe recent times has changed the equations. The Performance of the Database plays a vitalrole in the competition. In this research, an attempt has been made to evaluate and comparethe performance of the traditional database and the Cloud Database.Objectives: This thesis investigates about the prior works on the issues that affect theperformance of Cloud Database. And compares the performance of a Database in Traditionalto that Cloud EnvironmentsMethods: Two different research methods are used to carry the research. They areSystematic Literature Review (SLR) and Quantitative Methodology. Articles from ScientificDatabases are chosen for SLR process.Results: From the SLR process, 4 issues were identified. From the Experimentation results,Cloud Database is having poor performance compared to the Traditional Database.Conclusions: Issues that affect the performance of Cloud Database are identified and a testbed is created to test the performance of a Database. Attempts are to be made to improve theperformance of Cloud Database.Keywords: Database, Cloud Computing, Performance, affects

ACKNOWLEDGMENTAny attempt at any level cannot be satisfactorily completed without the supportand guidance of our Supervisor. We express heartfelt gratitude to Prof. LarsLundberg for his immense support to carry out this work. We are much thankfulto librarian Sophia Swartz for her guidance in SLR. We are greatly thankful to ourbeloved parents, brothers and friends for their relentless support that they hadgiven us to reach our goals.Yours truly,Kaushik Donkena,Subbarayudu Gannamani.ii

CONTENTSABSTRACT .IACKNOWLEDGMENT. IICONTENTS .IIILIST OF FIGURES . 1LIST OF TABLES . 2LIST OF ABBREVIATIONS . 31INTRODUCTION . 41.11.21.32AIMS AND OBJECTIVES . 4RESEARCH QUESTIONS . 5THESIS OUTLINE . 5BACKGROUND . 62.1DATABASE . 82.1.1 Database Management System . 92.1.2 Database Optimization. 102.2CLOUD COMPUTING . 122.3 DEPLOYMENT MODELS . 132.2.1 Private Cloud . 132.2.2 Community Cloud: . 132.2.3 Public Cloud. 142.2.4 Hybrid Cloud . 142.3SERVICE MODELS . 142.3.1 Software as a Service (SaaS) . 142.3.2 Platform as a Service (PaaS). 142.3.3 Infrastructure as a Service (IaaS) . 143RESEARCH METHODOLOGY . 153.1SYSTEMATIC LITERATURE REVIEW (SLR) . 153.1.1 Planning the review . 163.1.2 Conducting the review. 173.1.3 Identification of Research . 173.1.4 Study Selection Criteria . 183.2EXPERIMENT . 193.2.1 On Traditional Database. 193.2.2 Constructing a test bed. 213.2.3 Database Normalization . 223.3CLOUD DATABASE . 224RESULTS . 244.1SLR RESULTS . 244.2EXPERIMENTAL RESULTS . 244.2.1 QUERY 1 . 254.2.2 QUERY 2 . 264.2.3 QUERY 3 (SELECT COMMAND USING SIMPLE JOIN). 274.2.4 QUERY 4 (SELECT COMMAND USING COMPLEX JOIN) . 295DISCUSSION . 315.1VALIDITY THREATS . 31iii

5.1.15.1.25.1.35.1.46Construct Validity . 32Internal Validity. 32External Validity . 32Conclusion Validity. 33CONCLUSIONS . 346.1LINKING RESEARCH QUESTIONS . 346.1.1 Research Question 1 . 346.1.2 Research Question 2 . 346.2FUTURE WORK . 34REFERENCES . 35APPENDIX A . 38APPENDIX B. 39APPENDIX C . 40APPENDIX D . 41APPENDIX E. 42APPENDIX F . 43APPENDIX G . 44APPENDIX H . 45APPENDIX I . 46APPENDIX J . 47iv

LIST OF FIGURESFigure 2-1 Journey of Relational Database Management System . 7Figure 2-2 Cloud Database as a Service . 8Figure 2-3 Database. 9Figure 2-4 Database Management System . 10Figure 2-5 Database Performance Optimization Dependency levels . 11Figure 2-6 Cloud Usage . 12Figure 3-1 showing the entity relationship diagrams for Employee database. 20Figure 3-2 Database schema of EMPLOYEE Database . 21Figure 4-1 Slow Down Factor between Traditional and Cloud Databases for different entriesfor Query 1 . 26Figure 4-2 Slow Down Factor between Traditional and Cloud Databases for different entitiesfor Query 2 . 27Figure 4-3 Slow down Factor between Traditional and Cloud Databases for different entitiesfor Query 3 . 28Figure 4-4 Slow Down Factor between Traditional and Cloud Databases for different entitiesfor Query 4 . 301

LIST OF TABLESTable 2-1 Advantages and Disadvantages of using Indexes . 12Table 3-1 Research plan . 15Table 3-2: Defining Research Questions . 16Table 3-3 Quality Assessment checklist . 17Table 3-4 Selection Criteria . 17Table 3-5 SLR Process . 19Table 3-6 Entities and attributes in Employee database . 21Table 3-7 Entity relationship and keys information . 22Table 4-1 SLR Results . 24Table 4-2 Query 1 Response Time Values of different entries for Traditional and CloudDatabase in milliseconds . 25Table 4-3 Data entries of the Query 1 . 25Table 4-4 Query 2 Response Time Values of different entries for Traditional and CloudDatabase in milliseconds . 27Table 4-5 Data entries of the Query 2 . 27Table 4-6 Query 3 Response Time Values of different entries for Traditional and CloudDatabase in milliseconds . 28Table 4-7 Data entries of the Query 3 . 28Table 4-8 Query 4 Response Time Values of different entries for Traditional and CloudDatabase in milliseconds . 29Table 4-9 Data entries of the Query 4 . 292

LIST OF base Management SystemDatabase as a ServiceData Manipulation LanguageInfrastructure as a ServicePlatform as a ServiceSoftware as a ServiceStructured Query LanguageSystematic Literature ReviewRelational Database Management System3

1INTRODUCTIONA Cloud can be defined as a parallel and distributed system which has a number ofvirtualized and interconnected computers. These are actively provisioned andpresented as single or more united computing resources depending upon the servicelevel agreement. Cloud has three popular computing paradigms Infrastructure as aService (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS). Theseservices include distributed operating system, the distributed database and otherservices.The Cloud Computing database is required apace and effectively and shouldreduce the burdens during routing configuration. The Cloud Database is constructed bycollecting a number of sites. The sites are also called as nodes which are interlinked bya communication network. Every single node is a database class. Each database classhas its own database, terminals, the central processor and their individual localdatabase management system.A database is an organized collection of data. A Database Management System(DBMS) is a software package with computer programs that controls the creation,maintenance, and use of a database. It allows the organizations to convenientlydevelop databases for various applications. A database is an integrated collection ofdata records, files and other objects. A DBMS allows different user applicationprograms to concurrently access the same database. DBMSs may use a variety ofdatabase models, such as the relational model or object model to conveniently describeand support applications. The term database is correctly applied to the data and theirsupporting data structures, and not to the database management system. The databasealong with DBMS is collectively called Database System.A Cloud Database is a database that typically runs on a Cloud Computingplatform, such as Windows Azure, Amazon EC2, GoGrid and Rackspace. There aretwo common deployment models: users can run databases on the cloud independently,using a virtual machine image, or they can purchase access to a database service,maintained by a Cloud Database provider. .Of the databases available on the Cloud,some are SQL-based and some use a NoSQL data model.1.1Aims and objectivesAim of the thesis is to evaluate the performance comparisons of traditional andnormal database and open doors for research on the performance issues in CloudDatabase. Creating and deploying data into the traditional databaseMigrating and deploying data into Cloud DatabaseTest traditional database performanceTest Cloud Database performanceCompare the results of traditional database and Cloud Database in termsof response time4

1.2Research questions1. What are the issues that affect the performance of a Cloud Database?2. What is the performance in terms of response time of a Cloud Database compared totraditional database?1.3Thesis OutlineIntroduction part describes the brief introduction to the research work. Backgroundconsists of background of Databases and the background of Cloud Computing.Research Methodology discusses the methodologies used for the research. Thisconsists of SLR and Quantitative Methodology. Results chapter presents theSLR(Systematic Literature Review) Results and Experimentation results. Discussiongives a brief discussion on the obtained results. Conclusions chapter discusses theconclusions linking the research questions and the future directions of the research.References give the list of used citations and Appendix gives information on theexperiment and its results.5

2BACKGROUNDThe concept of database management system is quite interesting to look at over aparticular period of time. According to [27], Database Management is developed infour phases from 1970’s to late 1990’s. Figure [1] clearly illustrates four phases ofDatabase Management System. In early 1970’s, organizations used IBM’s informationmanagement system (IMS) which stores the data using hierarchical model. But theorganizations have to maintain expensive main frames in order to relay on IBM’s IMS.By early 1980’s, IBM’s IMS is replaced by the Relational Database ManagementSystem (RDMS) such as Oracle. In 1980’s and 1990’s amplification of networkingDBMS technology is allowed on personal computers. After RDBMS progress to client/server environments and it’s implemented on large organizations. In 1990’s becauseof the fast growth of the technology symmetric multiprocessing system and datawarehousing options are made available on the RDBMS.6

Figure 2-1 Journey of Relational Database Management SystemAccoring to [29] Figure 2-1 shows the phases of the Relational DatabaseManagement System. This has kept growing and now this time it shifted to otherdimension i.e Cloud Computing. Cloud Computing has been an interesting paradigmin the recent times due to its advantages like scalability, virtualization and pay per use.As pay per use is involved, it is important to consider the resource utilization. CloudComputing is more helpful for IT industries to improve the management of their ownresources in an easy manner. Cloud Computing provides different services such asInfrastructure-as-a-Service(SaaS), Platform-as-a-Service(PaaS) and Software-as-aService(SaaS). According to [33] there is an addition to this list of services, calledDatabase-as-a-Service(DaaS). In this service, organizations host their own databases in7

Cloud Computing. This service provides the acess for DML(Data ManipulationLanguage) statement features (strore, retrieve, update and delete the data) via theinternet following [29].Figure 2-2 Cloud Database as a ServiceAccording to [28], a Cloud Database is a combination of different number ofnodes (or site collections) and each node has its own database, linked together in thecommunication network. Cloud Database system is a novel trend in the researchbecause many organizations want to migrate their databases into Cloud to exploit thebenefits Cloud Computing. Organizations look at the performance factor of thedatabases regardless of the paradigm, whether traditional or Cloud. In [30], authorsconducted various experiments on On-premisis traditional database in terms of IBM’SDB2, Oracle database and Microsoft SQL Server. The performance of the CloudDatabase is evaluated in this research and a comparison is made with that of an onpremises traditional database.2.1DatabaseDatabase is a collection of data or information in a well-organized manner so thatdata can be accessed, updated and managed easily. It can be imagined as a large datafile storing the data as in the following.8

Figure 2-3 DatabaseAs shown in the figure database is an integrated collection of data items or files.According to [31], the authors suggested the databases have to support features such ashigh reliability, high availability, high throughput and security. A database is rated as ahigh quality database if it supports aforementioned features in all operations suchupdating, managing, and retrieving of data. Enterprises will plan for the provision ofthese features while providing service to the database users.2.1.1Database Management SystemA Database Management System is software with computer programs that lets theuser control the creation, maintenance, and use of a database. According to [32]database package provides to the user a database engine, a data dictionary and a userinterface. The database engine is used for the purpose of effective storage and retrievalof data. The purpose of user interface is to create a new database or update an existingdatabase in the system. According to IBM dictionary of computing a data dictionary isa centralized repository of information about data such as meaning, relationships toother data, origin, usage, and format. It is a document which determines the structureof a database and describes a database. A DBMS can facilitate the concurrent access ofmultiple databases via user interface.9

Figure 2-4 Database Management SystemAccording to [31], a Database Management Systems acts like a platform fordatabase administrators to manage, create and update the database. Users can runcertain applications in the DBMS to access, modify and update the data. According to[32] there are different kinds of databases such as network, hierarchical and relational.Relational database was proposed by E.F.Codd in 1963. A relational database is thepredominant choice in storing data, over other models like the hierarchical databasemodel or the network model.2.1.2Database OptimizationAccording to [4] enterprises are becoming data-centric and increasingly producinghumongous amounts of data in the form of sales, retail records and other commercialinformation. This data stored in the database needs to be effectively managed.Enterprises analyze these databases continuously and take informed decisions based onthe analysis, so database performance plays a vital role in the overall functioning of thedatabase. At the time of creation of database the scale of meta-data related to thedatabase is small. As the size of the database increases, it encounters gradualdeterioration in the performance. This performance degradation motivated theresearchers to search for ways to improve the performance by database optimization.Database optimization can be performed at four different layers as shown in Figure 25.10

Figure 2-5 Database Performance Optimization Dependency levelsIn these four levels top most level is the SQL application level optimization. In thisoptimization the transaction time is reduced by indexing the database thereby leadingto improvement in the performance. The database performance translates to reductionin CPU costs in [35]. By indexing the database, the DBMS is enabled to maintain aseparate database object storing the metadata related to database. These objectscontained a sorted list of column values which contains row identifiers to thecorresponding rows in that table as shown in [34].Indexes are internally organized in a tree structure. According to [37] there arecertain disadvantages of using the indexes to the database. Usage of the indexes resultsin speed up in the query execution, retrieval of data but every additional index addedto the index table slows down the manipulation further. Since everyINSERT/DELETE/UPDATE can be processed only after updating all thecorresponding indexes it takes additional CPU cycles and time to keep the indexessynchronized with the tables. This also results in Database consuming additional spacein database.11

Table 2-1 Advantages and Disadvantages of using nce2Using indexes we can speed upqueriesMaintenance overhead3Reduce CPUexecutionIndexes occupy the additional space indatabasecostdatabaseforquery452.2Using Index slows down manipulationfurtherAvoids full table scan in searchqueriesINSERT/DELETE/UPDATE can beprocessed only after updating all thecorresponding indexesTable data can be stored in anorganized wayNeed to maintain index and tablesynchronization every time.Cloud ComputingIt is hard to define what Cloud Computing is because different authors havedifferent definitions on Cloud Computing. But according to NIST (National Instituteof standards and technology)“Cloud Computing is a model for enabling ubiquitous,convenient, on-demand network access to shared pool of configurable computingresources (e.g., networks, servers, storage and applications) that can be rapidlyprovisioned and released with minimal management effort or service providerinteraction”.Figure 2-6 Cloud UsageCloud Computing has five essential characteristics (On-demand self-service,Broad network access, Resource pooling, Rapid elasticity and Measured service), three12

service models (Software as a service, Platform as a service and Infrastructure as aservice) and four deployment models (Private Cloud, community Cloud, public Cloudand Hybrid Cloud).2.3 Deployment ModelsAccording to [36] four types of deployment services available in the Cloud theyare Private Cloud, Public Cloud, Hybrid Cloud, and Community Cloud. Below Figure2-7 Cloud Deployment Models clearly illustratesFigure 2-7 Cloud Deployment ModelsThe above figure clearly shows the variation between the private, Public, andHybrid Clouds. Company ‘A’ owns private Cloud whereas company ‘B’ and company‘C’ owns Public Cloud.2.2.1Private CloudPrivate Cloud is also called as internal Cloud or corporate Cloud. Private Cloud isproviding resource, storage of data to a limited number of hosted services. This Cloudmay be managed and operated by the organization behind a firewall. Private Cloud canaccess who are positioned within the boundaries of an organization.2.2.2Community Cloud:Community Cloud is a type of infrastructure to share a resource to manyorganizations from a specific community with common concerns (e.g. securityrequirements, mission, policy, compliance considerations).13

2.2.3Public CloudThis cloud infrastructure is employed for delivering resources to general publicover the internet for open use. It may be managed and owned by academia foracademic purposes or by the government or corporate for commercial purposes.2.2.4Hybrid CloudThis cloud infrastructure is a combination of two or more distinct clouds. In thismodel an organization provides and manages some resources in-house and has othersprovided externally. It offers the benefits of multiple deployment models to the users.2.3Service models2.3.1Software as a Service (SaaS)“This capability provided to the consumer is to use the provider’s applications running on aCloud infrastructure. The applications are accessible from various client devices througheither a thin client interface, such as a web browser (e.g., web-based email), or a programinterface.” [26]2.3.2Platform as a Service (PaaS)“This capability provided to the consumer is to deploy onto the Cloud infrastructureconsumer-created or acquired applications created using programming languages, libraries,services, and tools supported by the provider.” [26]2.3.3Infrastructure as a Service (IaaS)“This capability provided to the consumer is provision processing, storage, networks, andother fundamental computing resources where the consumer is able to deploy and runarbitrary software, which can include operating systems and applications.” [26]14

3RESEARCH METHODOLOGYThe two research questions follow two different methodologies. They arerepresented in th

The term database is correctly applied to the data and their supporting data structures, and not to the database management system. The database along with DBMS is collectively called Database System. A Cloud Database is a database that typically runs on a Cloud Computing platform, such as Windows Azure, Amazon EC2, GoGrid and Rackspace.

Related Documents:

Getting Started with Database Classic Cloud Service. About Oracle Database Classic Cloud Service1-1. About Database Classic Cloud Service Database Deployments1-2. Oracle Database Software Release1-3. Oracle Database Software Edition1-3. Oracle Database Type1-4. Computing Power1-5. Database Storage1-5. Automatic Backup Configuration1-6

sites cloud mobile cloud social network iot cloud developer cloud java cloud node.js cloud app builder cloud cloud ng cloud cs oud database cloudinfrastructureexadata cloud database backup cloud block storage object storage compute nosql

Cloud Database Battle: AWS vs. DIY vs. Oracle Cloud Database Premises The first premise of this research is that architecting the Oracle Cloud Database service to run on specialized hardware and software, either on-premises or in Oracle Cloud Infrastructure (OCI), allows the cloud database vendor to reduce costs significantly.

Database In-Memory is an option for Oracle Database Enterprise Edition Database In-Memory was included in the first patchset (12.1.0.2) for 12.1 and all subsequent Oracle Database releases Available: - Database Cloud Service - Virtual Machines: Extreme Performance - Database Cloud Service - Bare Metal: Extreme Performance

Why you need to upgrade to Oracle Database 19c? . Public Cloud DBA 2015 Oracle Public Cloud with Database Cloud Service, Database Backup Cloud Service, Storage Cloud Service, IaaS Cloud Service PDBAs 2017 Multi-Tenant with O

Database Applications and SQL 12 The DBMS 15 The Database 16 Personal Versus Enterprise-Class Database Systems 18 What Is Microsoft Access? 18 What Is an Enterprise-Class Database System? 19 Database Design 21 Database Design from Existing Data 21 Database Design for New Systems Development 23 Database Redesign 23

Oracle Cloud Management Pack for Oracle Database delivers capabilities spanning the entire Database cloud lifecycle. It lets Cloud administrators identify pooled resources, configure role-based access, define the service catalog, and the related chargeback plans. It allows Cloud users to request database services, and consume them on

The Adventure Tourism Development Index (ATDI) is a joint initiative of The George Washington University and The Adventure Travel Trade Association (ATTA). The ATDI offers a ranking of countries around the world based on principles of sustainable adventure tourism