(IJCSIS) International Journal of Computer Science and Information Security, Vol. 11, No.

6, June 2013

Toward a New Approach for Modeling Dependability of Data Warehouse System
Imane Hilal
RITM Lab., CED Engineering Sciences ENSEM ESTC, Hassan II University Casablanca, Morocco

Nadia Afifi
RITM Lab., Computer Engineering Department ESTC, Hassan II University Casablanca, Morocco

Reda Filali Hilali
RITM Lab., Computer Engineering Department ESTC, Hassan II University Casablanca, Morocco

Mohammed Ouzzif
RITM Lab., Computer Engineering Department ESTC, Hassan II University Casablanca, Morocco

Abstract—The sustainability of any Data Warehouse System (DWS) is closely correlated with user satisfaction. Therefore, analysts, designers and developers focused more on achieving all its functionality, without considering others kinds of requirement such as dependability’s aspects. Moreover, these latter are often considered as properties of the system that will must be checked and corrected once the project is completed. The practice of "fix it later" can cause the obsolescence of the entire Data Warehouse System. Therefore, it requires the adoption of a methodology that will ensure the integration of aspects of dependability since the early stages of project DWS. In this paper, we first define the concepts related to dependability of DWS. Then we present our approach inspired from the MDA (Model Driven Architecture) approach to model dependability’s aspects namely: availability, reliability, maintainability and security, taking into account their interaction. Keywords-component; Data Warehouse System; Model Driven Architecture ; Dependability; Availability; Reliability, Security, Maintainability.

The final goals to implement the DWS are: (i) To evaluate complex queries without causing severe impact on the sources; (ii) To increase data availability and decrease response-time for OLAP queries; (iii) To provide correct historical trends for state-oriented data; (iv) To protect and secure the crucial business information; (v) To back up modifications and insure evolution and maintenance. Those goals involve the guarantee of non-functional requirements such as availability, reliability, security, confidentiality, integrity and maintainability which are encompassed in the dependability’s attributes [3]. To meet those goals, we suggest an approach that spans over dependability’s attributes. Our contribution presents, on the one hand, the advantage of considering these attributes from the early stages of the DW project. On the other hand, it provides models to the designers and developers in order to realize these functionalities in respect to the Model Driving Architecture’s (MDA) principles. The remainder of this paper is organized as follows: section 2 will present the related work on the development of dependable DWS; section 3 will give an overview of DW, dependability aspects and MDA; section 4 will introduce our approach through which we develop our models for DWS dependability. An example of implementation is shown in section 5. Finally, in section 6 we present our conclusion and future work. II. RELATED WORK

I.

INTRODUCTION

Data Warehouse Systems (DWS) are specially used by decision makers to analyze the status and the development of their organization [1], based on a large amount of enterprise information integrated from heterogeneous Data Sources [2]. This information is organized following a multidimensional structure to make exploration and analysis easier for users. The DWS’s architecture is composed from several layers: (i) Heterogeneous Data Sources (DS), (ii) ETL (Extraction/ Transformation/ Loading) process which extract and transform data from these DS, and load the information into DW ,(iii) Data Warehouse repository, (iv) Restitution Tools that analyze the data in OLAP way.

In DW’s literature, dependability’s attributes have been neglected or they have been presented as a second class type of requirements, and have also been considered as the other non-functional requirements. But the experiences show that capturing non-functional requirements without mapping them into the conceptual model may provoke loss of information [4]. Only few works have specifically addressed this issue. In

47

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

( (IJCSIS) Interna ational Journal of o Computer Scie ence and Information Security, o. 6, June 2013 Vol. 11, No

par rticular, Pain &castro & [5] pr roposed (DWA ARF) the Dat ta Wa arehouse Requ uirements Defin nition approach h that adapts a req quirement engin neering proces ss for requirem ments definition n and d management of DW. Their approach dem mands particula ar atte ention to non-functional req quirements tha at are captured thro ough an exten nsion of the NFR framewo ork. The sam me auth hors have pro ovided an exha austive classifi fication of non nfun nctional requir rements that must be add dressed in th he dev velopment of DWS D [6]. A different pr roposal comes from V. Peralt ta, A.Illarze, R. R Rug ggia, [7] who have h suggested d modeling the e non-functiona al req quirements thro ough guidelines that are not directly related wit th the conc ceptual mode el. Instead, non-functiona al req quirements hav ve been used during d logical design, wher re most of the choic ces related to the performance and security y ve already been n taken. hav E. Soler, V. Stefanov, J. N. Mazón, J. Trujillo, E. E Fer rnández-Medin na, M. Piattin ni, [8] propos sed a security y req quirement mode el for DW bas sed on the MD DA approach[9] ]. Thi is approach wa as extended by y Carlos Blanco [10] who ha as dev veloped a secu ure DW, inclu uding security issues in each h stag ge of the deve elopment proce ess. This is the e only approach h that t considers sec curity as a non n-functional req quirement from m the early stages of f the DW devel lopment. On the other hand, h some asp pects of depend dability such as a reli iability, availab bility, security and maintainab bility have been n wid dely discussed in the DWS[11]. However, their t interaction n and d integration we ere rarely discu ussed [11, 12]. III. DEPEN NDABILITY’S S ATTRIBUTE ES OF DATA WAREHOUS SE SYSTEM

from the DW us sing, for insta ance, query tools. components of a typ pical DWS are shown in “Fig. .1”.

The

Figure1.D Data Warehouse Sy ystem’s architectur re

C. DWS’s dependability In n this section we w will adapt each e dependab bility attribute in the e DWS context: 1) Availability m to deliver Availability is the readiness of the system ct service whe enever solicited d [13]. The sam me definition correc is pro ojectable on DW WS. Thus, we e can define the availability of th he DWS as th he latter’s abi ility to provid de extracted, cleane ed, rebuilt inf formation whe enever needed d, and which can be analyzable fo ollowing differ rent axes. iterature provid des many optim mization techn niques to deal Li with the demand of o availability y in DWS. Most M of these n DW, and are e inherited from traditional techniques focus on onal database systems. s Amon ng these techni iques [14] we relatio find: materialized views, v indexing g methods, data a partitioning p [16 6]. Regarding ETL, many [15], and parallel processing en developed to o improve its availability a as techniques have bee m responsi ible for transfo orming and del livering data. it’s mainly Those e techniques have propose ed to optimiz ze the ETL proce ess, using the workflow w conce epts [17] and th he concept of near-r real-time data warehousing w [18, 19]. 2) Reliability he common Reliability’s R d definition con nsists of the Th contin nuity of correc ct service [13] ]. Projecting th his definition to DW WS allows us to t consider this s attribute as th he percentage of tim me the DWS is available for f use. While e considering aspec cts of maturity, fault tolerance e and recoverab bility. Co onsidering the definitions of f availability an nd reliability, both emphasize th he avoidance of failures, and can be ped together, and a collectively y defined as th he avoidance group or min nimization of service s outages s. 3) Maintainabil lity M Maintainability t bear repairs s and updates is the ability to [13]. In reality, DWS D are int teracting with a dynamic onment due to o changing bu usiness rules or o operational enviro object tives on one hand, and increasing requi irements and chang ging in its data a sources on the t other hand d. Thus, their maint tenance is ver ry important to insure the eir ability of adaption to their env vironment’s evolutions.

A. Dependability y’s attributes The original definition d of dependability d fo a computing for g system is the abi ility to deliver service that can c legitimately y be trusted [13]. . This definition induces the need fo or tification of tru ust to avoid ser rvice failure an nd make it mor re just acc ceptable. y may be viewed according to t different, bu ut Dependability com mplementary attributes, a wh hich are summ marized in: (i i) Availability: abil lity of the system to be ope erational at th he me requested (ii) ( Reliability: continuity of o service, (iii i) tim Sec curity: non-oc ccurrence of environmenta al catastrophic con nsequences, (iv) ( Confiden ntiality: the prohibition p o of una authorized disc closure of info formation, (v) Integrity: non nocc currence of improper i alter ration of inf formation, (vi i) Ma aintainability: the ability to o handle repai irs and update es [13 3]. Associating g integrity and availability to ogether, as wel ll as confidentiality, c , lead to securi ity. B. Data Warehou use System DWS are kn nown as a co ollection of de ecision suppor rt hnologies, aim ming at making g decisions be etter and faster r. tech ETL L (Extraction, , Transformati ion, Loading) retrieves dat ta from m various ope erational datab bases, cleans, transforms and load ds it in DW repository r [2]. The rules use ed to clean and tran nsform the data a and perform the transforma ation are part of o the metadata man nagement syst tem. Users can extract dat ta

48

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

( (IJCSIS) Interna ational Journal of o Computer Scie ence and Information Security, o. 6, June 2013 Vol. 11, No

Based on the e previous wo ork, we have to distinguish h betw ween two conc cepts: • Maintaina ability: which is i the ability of o an element to o be maint tained; this ab bility stems fr rom all design n features th hat enhance the e ability of serv vice. Maintenan nce: is a seri ies of actions of appropriat te character (content, timin ng, quality) to restore or keep p nt in an operati ional state. an elemen

he requiremen nts of the futur re system are described in Th the CIM, C which is refined into the PIM. The PSM is the result t of the PIM transformation n. The process refines the PIM based on the specification described in the Platform ription Module e (PDM) defin ning how to use u a specific Descr platfo orm. Th he main advan ntages [21] of th he MDA frame ework are: (i) Resul lts are automa atically genera ated which is expected to impro ove productivi ity, developme ent duration, and a cost; (ii) The developer d pay y more attentio on to CIM and PIM instead of foc cus on logical and a technical details; d (iii) PIM M is portable to dif fferent target platforms; p (iv) Once a transf formation has been developed, it t can be reus sed whenever needed; (v) ges have to be b done in th he PIM only if the target Chang platfo orm has change ed; (vi) New re equirements in n the CIM are passed from PIM to PSM imm mediately and changes are cted automatica ally [14], [15], [16]. reflec V V. OUR APP PROACH

Based on the definitions abo ove, the two concepts are no ot con ntradictory, since maintenance is the action of maintaining g. Mo ore the system accepts maint tenance’s operation, the mor re it’s maintainable. In our propos sal, we consid der that the two pects have the same s meaning. asp 4 Security 4) rement for div verse kinds of o Security is a basic requir app plications. Sin nce business information in the dat ta war rehouse is sen nsitive, the di ifferent aspect ts (availability y, inte egrity and co onfidentiality) of security are a particularly y imp portant. In this s way, informa ation security should s be taken n into o consideratio on from the e early stage es of DWS’ ’s dev velopment life-cycle [9, 10]. e, the integratio on of dependab bility aspects is i To summarize a necessary n to ins sure a dependa able DWS. Bu ut the challeng ge in this context is that the e attributes are a often: (i i) terogeneous; (ii) ( Ranked dif fferently and subjectively by y Het the stakeholders; (iii) Interrelate ed: in conflict or in harmony y. a the im mplementation n of these attrib butes requires a In addition, lot of expertise an nd domain know wledge. IV. MDA : MODEL DR RIVEN ARCHITECTURE

Our proposal aim ms to provide a dependable DWS D process n, taking into account, si ince the earl ly stages of design mode eling, dependab bility’s constraints. As thes se latter vary accor rding to the fie eld of use of DWS, D we sugg gest a generic appro oach aligned to the standa ard MDA, allowing their integr ration in a ref fined and incre emental way. Of course it based d on different le evels of abstra action to ensure e traceability, and check their compatibility while consi idering their action. “Fig. 3” ” explicitly repr resents our app proach. intera

MDA [20] pr rovides model-driven softwar re developmen nt bas sed on the sep paration of the e specification of the system m fun nctionality and d its implemen ntation. It defi fines models at a diff ferent abstracti ion levels, and d transformatio ons required to the passage of a le evel to another [21] “Fig. 2”. This approa ach propose business models m (CIM M: Com mputation In ndependent Model) M based d on system m req quirements. Conceptual C models (PI IM: Platform m Ind dependent Mod del), which do not n include inf formation abou ut spe ecific platforms s and technolog gies. These latter are including in the logical models m (PSM: Platform Sp pecific Model) ). veloping inform mation systems s applying the MDA approach h Dev imp proves product tivity, saving time t and effor rt, and provide es sup pport for syste em evolution, integration, interoperability i y, por rtability, adapta ability and reus sability [22].

Fig gure 3. Our propos sed approach

CIM
Co omputati on nIndepen dant Model

PIM
Platform Independe nt Model
Transformation

PSM
Platform Specific Model

c MDA Models Figure 2. Generic

Our approach covers the entir re DWS project. However, D is itself a heterogeneo ous system, be ecause of the the DWS platfo orms from wh hich it is co omposed (DS, ETL, DW, analysis tools), we adapted the "d divide to conqu uer" strategy. ne hand, it allo ows to deal with each layer preserving p its On on own characteristics s, and secondl ly integrate th hem with the unding layers s. We have thus split the t standard surrou archit tecture of DWS into four ho omogeneous lay yers: the DS, ETL, DW and analysis a tools. . Then we conducted c a tative study to list all parame eters influencin ng the aspects qualit

49

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 11, No. 6, June 2013

of dependability. We have integrated them through building a structured approach respecting the different types of models proposed by the MDA approach. A. Requirement Level This section is devoted to the qualitative study of all dependability’s attributes in the four layers of the DWS. We isolated each of its layers. Based on the fact that the availability and reliability are very close (see section 3). They will therefore be associated in the following of our work. We can list the parameters that influence dependability’s attributes considering that their implementation depends on the type of their attribution. 1) Data Source (DS): Availability and Reliability <Structuration, Volumetric, Optimization, Data Quality, Support, Replication> as: •Structuration <structured, mi structured, not structured> •Volumetric <transaction frequency, recording number > •Optimization <index, fragmentation, view, parallelism > •Data Quality mistaken> <redundant, missing, incomplete,

•Metadata <source, destination, cleaning operations, management rules, constraints> •Modularity <interdependence transformations, subsystems decomposition > •Complexity <calculation rules, DS number> •Restoration <recovery management, exception management> Security: <Intrusion Management, Audit, Restoration> as: •Intrusion Management <firewall, encryption, codification, access control > •Audit <data, transformation> •Restoration <recovery management, exception management> 3) Data Warehouse (DW) Repository Availability and Reliability <Schema, Volumetric, Complexity, Replication, Optimization, Support, loading Frequency> as: •Schema <star, snowflake, constellation> •Volumetric <recording number, access frequency, granularity> •Complexity < fact number, dimension number, hierarchy number> •Optimization <index, fragmentation, view, parallelism> •Support <hard disc, communication> Maintainability <Documentation, Architecture, Complexity, Optimization, Support> as: •Documentation <metadata, schema, dimension> •Architecture <centralized, distributed, consolidation Data Mart> •Complexity <dependence dimensions, granularity> •Optimization <index, fragmentation, view, parallelism> •Support <hard disc, warm maintenance, cold maintenance > Security <User Management, Data Management, Support> as: •User Management <authentification, permission, quotas, audit, hierarchy> •Data Management <integrity, authenticity, completeness, locking> •Support <Firewall, encryption, replication> 4) Restitution tools Availability and Reliability <Type, Volumetric, Complexity, Support> as: •Type < Ad hoc query, parameterizable query, dashboard, cube OLAP, Data Mining> •Volumetric <data size> •Complexity <axis number, transformation, granularity> •Support <RAM, Communication> Maintainability <Type, Complexity> as: Documentation, Modularity,

•Support <hard disc, communication> Maintainability: <Documentation, Optimization, Support> as: •Documentation <schema, metadata> •Structuration <structured, mi structured, not structured> •Optimization <index, fragmentation, view, parallelism > •Support <hard disc, communication> Security: <User Management, Data, Support> as: •User Management< authentification, permission, quotas, audit, hierarchy> •Data <integrity, authenticity, completeness, locking> •Support <firewall, encryption, replication> 2) Extraction, Transformation, Loading (ETL) Availability and Reliability <Data Quality, Periodicity, Cleaning, Complexity, Support> as •Data Quality mistaken> <redundant, missing, incomplete, Structuration,

•Periodicity <batch, real time, cycle refreshment> •Cleaning <consolidation, formatting, restructuration, normalization, doubloon elimination, aberrant values elimination, missing values > •Complexity < data size, calculation rules, DS number> •Support <RAM, temporary storage, communication> Maintainability: <Metadata, Modularity, Complexity, Restoration> as:

50

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

( (IJCSIS) Interna ational Journal of o Computer Scie ence and Information Security, o. 6, June 2013 Vol. 11, No

•Type < Ad A hoc quer ry, parameter rizable dashboard d, cube OLAP, , Data Mining> >

query y,

•Documentation <metadata, application ns, indicators s, on rules> calculatio •Modularity <interdependen < nce, parallelism m> •Complexity <Axis < number, , transformation, granularity> > Security: < User Manag gement, Data Management t, Sup pport> as: ement<authent tification, perm mission, quotas s, •User Manage audit, hier rarchy> •Data Manage ement <integrit ty, validity, pri ivacy> •Support <fire ewall, encryption> B. Computation Independent I Level At this stage e, we will implement i the e Computation n dependent Mod del (CIM) usin ng the NFR (N Non Functiona al Ind Req quirements) Fr ramework appr roach. This mo odel is based on n the parameters fro om the previou us level. The NFR N Framework k pro oposed by Ch hung [23], is based on th he concept of o "So oftgoal" to re epresent infor rmation about t the differen nt qua alities of a syste em, and their in nterdependenci ies [24]. The develop pment of interactive SIG S (Softgoa al Inte erdependency allows repr resenting th he Graphs) dep pendability’s at ttributes and their t integration. Then, it can n exa amine the inter ractions and id dentify potenti ial conflicts, to bett ter suggest a compromise c in n the early stag ges of the DWS S proj oject. The follow wing diagrams s “Fig. 4, 5, 6” clarify the CIM M ava ailability & re eliability, secu urity and mai intainability of o DW WS.

Figure 5. Maintaina ability CIM

Figure 6. Securi ity CIM

Fig.7” shows an a example of integration of o the above “F mode els (“Fig. 4, 5, 5 6”). This integration i wi ill allow the design ner to view all a the dependa ability ’ s con nstraints in a mode el. Therefore, the appropr riate impleme entation ’ s decisi ions of functio onal modules will be made e taking into accou unt non-fun nctional requ uirements r relating to depen ndability’s one. .
Figure 4. Availability A & Reli iability CIM

51

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

( (IJCSIS) Interna ational Journal of o Computer Scie ence and Information Security, o. 6, June 2013 Vol. 11, No

In this par rt we relied on n an existing UML U profile [27]: UML profile for f Quality of Service "QoS" " to represent the properties of quality of o service that t include ndability. “Fig g. 9, 10, 11”represent the t PIM of depen availa ability & reliability, maintainability and a security accor rding the profile already cited “Fig.12” shows an examp ple of their integration.

Fig gure7. Example of o CIM integration

C. Platform Indep ependent Level 1 The transit 1) tion from CIM to PIM To ensure trac ceability of dep pendability’s aspects since th he exp pression of req quirements to the implemen ntation, we wa as bas sed on the work w of S. Su upakkul, L.Ch hung [25]. Th he auth hors proposed d an integrat ted modeling g language by y extending UML with the NFR R framework using u the UML L pro ofile. They defi ined a MetaMo odel to represe ent the concept ts in the NFR Fram mework. They y alsoidentified d the extension n poi ints for associa ating the two no otations.
Figure 9. PIM‘s Availabi ility & reliability

Figure 8. MetaMo odel of NFR Fram mework & relations ship with UML metadata from fr [25]

“Fig. 8” expla ains the relation nship between the t Meta mode el of NFR N Framewo ork and the UM ML MetaModel l. Therefore, we w just tified the exis stence ofthe tr ransformation from our CIM M (usi ing NFR Fram mework) to PIM (using UM ML profile) by y exp ploiting this relationship [26].
M PIM M Figure 10. Maintainability

2 Developme 2) ent of PIM

52

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

( (IJCSIS) Interna ational Journal of o Computer Scie ence and Information Security, o. 6, June 2013 Vol. 11, No

However the DWS D is a sys stem composed of several atively differen nt layers. Thus, , for each layer r of the DWS opera and every e aspect of the dependability, a PS SM must be establ lishedto keep the t specificatio ons of each la ayer. To keep our ap pproach valid for all platform ms, we will do o not specify the PS SM because th hey usea specifi fic technology. Once the PSM M are refined, the generat tion code is matic. In addi ition, the integration of de ependability’s autom attribu utes with the functional f part t is guaranteed d through the aspec ct oriented prog gramming. In the t next sectio on we present a case e study to demo onstrate the va alidity of our ap pproach. VI. EXA AMPLE OF IMP PLEMENTATI ION

In n our case stud dy we take exa ample of imple ementation of some of aspects of f the Business Intelligence dependability, d e users require e as priority needs, n data ava ailability and where reliab bility. Howeve er, it suppos sed that data a are often maint tained. Thus to o ensure a depe endable DWS we w must take into account a the inte eraction betwee en the three asp pects. Ba ased on CIM and PIM prop posed by our approach, a we were able to select t the common parameters to o the various utes that are av vailability, reli iability and ma aintainability, attribu and on o which we ca an act. These parameters p are e: replication, suppo ort and optim mization (inde ex, views, fr ragmentation, parall lelism). Su upport and rep plication techn nics can be dir rectly applied on th he media platfo form of DWS. We chose to t implement fragm mentation in the DW rep pository. The tests were perfor rmed by a workload w query y running on n the APB-1 bench hmark [28] und der Oracle 10g g. We applied the t horizontal fragm mentation of 10 0 queries Q1…Q Q10 of thebenc chmark. The respon nse time comp parison of thes se queries befo fore and after fragm mentation is illu ustrated by “Fig g. 13”.

Figure 11. Se ecurity PIM

5000 4500 4000 3500

milllisecond

3000 2500 2000 1500 1000 500 0

with hout fragmenta ti on with h fragmenta ti on

Figure12. PIM M which integrate the dependability’ ’s attributes

Q Q3 Q4 Q5 5 Q6 Q7 Q8 Q9 Q10 Q1 Q2
Figure 13 3. Graphical repres sentation of results s

D. Specified Leve el Platform At this level, developers nee ed to combine information on n the PDM platform m (Platform Description D Mo odel) with PIM M g the PSM. P Given th he heterogeneit ty of platform ms to generate which composed DWS architecture, we suppo ose that specifi ic M must be dev veloped for eac ch of dependability attributes. . PSM

In n the following the summary of o results: 9 Fragmentati ion has improv ved the respons se time of the data wareho ouse which ha as a positive im mpact on the availability and reliability.

53

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

(IJCSIS) International Journal of Computer Science and Information Security, Vol. 11, No. 6, June 2013

9 Operations of maintainability and security can be slowed because of the complexity generated by the different fragments. 9 We can adjust the number of fragments in order to find a compromise between availability, reliability and maintainability, security taking into account the levels tolerated by users. VII. CONCLUSION AND PERSPECTIVE Our approach offers models at different levels of abstraction to deal with attributes of dependability. The main difficulty was to select measurable parameters common to the various aspects of the dependability. This problem has been overcome by our proposal, which integrates the different attributes taking into account their interaction from the early phases of modeling. To realize this, we used the models offered by the MDA. As perspective, we can propose the integration of aspectoriented programming approach in order to concretize the implementation of our proposal. We can also suggest the definition of an UML profile specific to dependability in DWS’s context. REFERENCES
R. Kimball,L.Reeves,M.Ross, andW.Thornthwaite, The Data Warehouse Lifecycle Toolkit, Wiley, 2004 [2] W.Inmon, Building the Data Warehouse, 3rd edition, Wiley, 2002. [3] A. Avizienis, J. Laprie, and B. Randell, “Fundamental concepts of dependability,” in Proc. 3rd Information Survivability Workshop, 2000, pp. 7–12. [4] M.Golfarelli,: From user requirements to conceptual design in data warehouse design: A survey. In: Bellatreche, L. (ed.) Data Warehousing Design and Advanced Engineering Applications: Methods for Complex Construction. IGI Global (2009) [5] F. R. S Paim, J. F. B Castro, (2003, September). DWARF: An approach for requirements definition and management of data warehouse systems. InRequirements Engineering Conference, 2003. Proceedings. 11th IEEE International (pp. 75-84). IEEE. [6] F. R. S.Paim,J. Castro, (2002, November) “Enhancing Data Warehouse Design with the NFR Framework”. In 5th Workshop on Requirements Engineering (WER2002), Valencia, Spain (pp. 40-57). [7] V.Peralta,A. Illarze, &R. Ruggia, (2003). On the applicability of rules to automate data warehouse logical design.In CAiSE Workshops. [8] E.Soler,V. Stefanov,J. N. Mazón,J.Trujillo, ,E. Fernández-Medina, , &M.Piattini, (2008, March). Towards comprehensive requirement analysis for data warehouses: Considering security requirements. . In: The Third International Conference on Availability, Reliability and Security ARES 2008, Barcelone, Espagne, pp. 104–111. IEEE Computer Society, Los Alamitos (2008) [9] E.Soler, R.Villarroel,J. Trujillo, E.Fernandez-Medina, ,M. &Piattini, (2006, April). Representing security and audit rules for data warehouses at the logical level by using the common warehouse metamodel. In Availability, Reliability and Security, 2006. ARES 2006. The First International Conference on (pp. 8-pp). IEEE. [10] C.Blanco, I.Garcia-Rodriguez de Guzman, E.Fernandez-Medina, , & J.Trujillo, (2010). Defining and transforming security rules in an MDA approach for DWs. International Journal of Business Intelligence and Data Mining, 5(2), 116-133. [1]

[11] L. Bellatreche, (2009). Data Warehousing Design and Advanced Engineering Applications: Methods for Complex Construction (Advances in Data Warehousing and Mining-Adwm. Information Science Reference-Imprint of: IGI Publishing. [12] E. Lau,&S.Madden, (2006, September). An integratedapproach to recovery and highavailability in an updatable, distributed data warehouse. InProceedings of the 32nd international conference on Very large data bases (pp. 703-714). VLDB Endowment. [13] AlgirdasAvizienis, Jean-Claude Laprie, Brian Randell, and Carl Landwehr,Basic Concepts and Taxonomy of Dependable and Secure ComputingIEEE Transactions On Dependable And Secure Computing, Vol. 1, No. 1, January-March 2004 [14] S.Chaudhuri,U.Dayal,&V. Narasayya, (2011). An overview of business intelligence technology. Communications of the ACM, 54(8), 88-98 [15] L.Bellatreche, M.Schneider, H. Lorinquer, &M.Mohania, (2004). Bringing together partitioning, materialized views and indexes to optimize performance of relational data warehouses. In Data Warehousing and Knowledge Discovery (pp. 15-25).Springer Berlin Heidelberg. [16] F.Dehne, T.Eavis, & A.Rau-Chaplin, (2003, May). Parallel multidimensional ROLAP indexing.In Cluster Computing and the Grid, 2003.Proceedings.CCGrid 2003. 3rd IEEE/ACM International Symposium on (pp. 86-93). IEEE. [17] A.Simitsis,P. Vassiliadis, &T.Sellis, (2005, April). Optimizing ETL processes in data warehouses. In Data Engineering, 2005.ICDE 2005.Proceedings. 21st International Conference on (pp. 564-575). IEEE. [18] T.Jörg,&S.Dessloch, (2010). Near real-time data warehousing using state-of-the-art ETL tools.In Enabling Real-Time Business Intelligence (pp. 100-117).Springer Berlin Heidelberg. [19] R. J.Santos, & J.Bernardino, (2008, September). Real-time data warehouse loading methodology. In Proceedings of the 2008 international symposium on Database engineering & applications (pp. 49-58).ACM. [20] D. S. Frankel, (2003). Model Driven Architecture ApplyingMda. John Wiley& Sons. [21] A. G.Kleppe,J. B. Warmer,&W.Bast, (2003). MDA explained, the model driven architecture: Practice and promise. Addison-Wesley Professional. [22] C.Blanco, I. Garcia-Rodriguez de Guzman, E. Fernandez-Medina, & J. Trujillo, (2010). Defining and transformingsecurityrules in an MDA approach for DWs. International Journal of Business Intelligence and Data Mining, 5(2), 116-133. [23] N. Subramanian, L. Chung, (2001, September). Software architecture adaptability: an NFR approach. In Proceedings of the 4th International Workshop on Principles of Software Evolution (pp. 52-61). ACM. [24] L. Chung, J. C. S. do Prado Leite, (2009). On non-functional requirements in software engineering. In Conceptual modeling: Foundations and applications(pp. 363-379). Springer Berlin Heidelberg. [25] S. Supakkul, L.Chung, (2005, August). A UML profile for goaloriented and use case-drivenrepresentation of NFRs and FRs. In Software Engineering Research, Management and Applications, 2005. Third ACIS International Conference on (pp. 112-119). IEEE. [26] M. R.Abid,D. Amyot, ,S. S. Somé, &G.Mussbacher, (2009). A UML profile for goal-orientedmodeling (pp. 133-148). Springer Berlin Heidelberg. [27] J. Ø.Aagedal, &E. F.Ecklund Jr, (2002). ModellingQoS: towardsa UML profile. In ‫ ا‬UML‫ ب‬2002—The UnifiedModelingLanguage (pp. 275-289). Springer Berlin Heidelberg. [28] Council, O. L. A. P. (1998). Apb-1 olap benchmark, release ii. http://www.olapcouncil. org/research/bmarkly.htm.

54

http://sites.google.com/site/ijcsis/ ISSN 1947-5500

Sign up to vote on this title
UsefulNot useful