SAP Business Intelligence Accelerator

Published on February 2017 | Categories: Documents | Downloads: 51 | Comments: 0 | Views: 359
of 8
Download PDF   Embed   Report

Comments

Content

 

I N F OR MAT M AT IO N FR AME AME WOR KS

THE SAP BUSINESS INTELLIGENCE  AC  A CCELE ELER RA T O R TURBO CHARGER FOR SAP NETWEAVER™ BUSINESS INTELLIGENCE

Naeem Hashmi Chief Strategy Officer Inf nfo orm rma ati tio on Frameworks May 19, 2006

 

Table of Contents Contents  The Bu Busin sine ess Intellig llige enc nce e Acc Acce ele lera rator... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ..... 3 How SAP Net NetWe Weaver ma makes Ana Analytic lytica al data ava vailab ilable le to its use users rs? ?.. .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... .... 4 HowSAP BI accele lera rator wor orks ks?...... ........... ........... ........... ........... ............ ........... ........... ........... ........... ............ ........... ........... ........... .......... ......... .... 4 How do SAP BI and BI acce accele lera rator work tog toga ather?... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ...... ..... 4  Ana  An alysis.......... ............ ............. ............ ............ ............ ............. ............ ............ ............. ............ ............ ............ ...... 5 Con onc clu lus sion ions s................................................. ......................................................................................................... ..................................................................... ............. 7  Ack  Ac knowledgments... ............. ............ ............ ............ ............. ............ ............ ............. ............ ............ ............ ...... 8

  Copyright Copyright ©2006, Inf nformation ormation Frameworks

2

 

THE BUSINESS INTELLIGENCE  AC  A CCEL ELER ERA ATO TOR  R  

Forr SAP NetWeaver™ Business Intell Fo ntelliigence T he s IIntelli ntellig ence e ge A Acce cce lera torS (A BI tor since ) is good news for SAP customers. Asi long lo ngBusines as I have beengenc enga d in the Paccelera BI design its con conceptio ception, n, I see what Sha  Agassi demonstrated at the SAP APP PHIR IRE E’06 in Or Orllando is a giant perfor formance leap for forward in end-user end-user access to the analytical data. One of the main pain poi points nts ffor or SAP NetWeaver BI (SAP BI) has been its data access perfo performa rmance due to it its s sophi sophisticated sticated relatio lational nal O OLA LAP P model. That That model works well ffor or small to mid sized business intel intelli lige gence operati tions. ons. No Now w with the BI accelerato tor, r, llarg arge SAP BI impl implem ementatio entations ns will enjo enjoy y super-high data access speed that is necessary to iimpleme mplement ‘‘rea reall-ti time me’ embedded-analytics lytics ffor or int intell elligent igent business solutio sol utions, ns, a key to SAP E Enterprise nterprise Services Archit Architectur ecture (ESA) based future business soluti sol utions. ons. To pro prove ve the super high perfo performa rmance gains, Shai demonstrated an analyti nalytical cal xA xApp pp that accessed llar argedatavolumes volumes in SAP BI thro throug ugh BI accelerator tor at a lightening speed. Shai stated in hi his s keynote: ‘…One Oneof of the thebig biggest iissu ssues in a data warehousing usingiis that you haveto aveto fi fig gureou ureout al alll tthe he queri rie es you are areg going oingto to get to to bui uilld cubes ahead of ttiime. You have aveto to al alm most tthi hink nk what  your users will think, and pre-think to design the architecture of a data warehouse. So what we[SA SAP] P] have havedo donew newith BI accelerat rator or iis s eff ffe ecti tive vely removed the thene need to buil build d the …. Y  Yo ou put the theda data, stored in the theBI BI accelerator, it stores iitt cubes… s….. N o mo more re Cubes…. in diffe ferent way than before, but iitt storesit on--lline, im immediia ate telyaccessiib blle e, and you don’t ….  So N o need to pre-think of any question that any user will ask …. Cubes…” Shai A gassi, Keynoteat teat Sahhpire’06, Orl Orlan ando, May17, 2006.

What caught my attenti ttention on werethe two key poin points ts emphasized during his keynote.  N  No ‘ o-More Cubes’ and ’ and “ youdon’t needtopre-think of anyquestionthat anyuser will ask”. Humm… ‘N ‘No o more Cubes’… perhaps this needs clarif rificatio ication n on what Shai meant which confused conf used quite a ffew ew SAP BI customers who approached me after his keynote. Let me explain why why suc such h conf confusion. usion. To best of my knowledge, BI accelerator tor is an ‘Appli ‘Applianc ance’ that speeds up Data Data Access (queries) task of SAP BI BI.. The BI accelerator sits on top (o (orr on the side) of a SAP BI instance inst ance. Data access is onl only y one import important ant piece of a data warehouse’s ov overa erall perf performanc ormance equatio ation. n. B BII accelerato torr addresses data access perfo performa rmance, scalabil labilit ity y and to so some me extent data load load-perfo performa rmance iissues ssues (as no aggregates are required anymore). Period.   Copyright Copyright ©2006, Inf nformation ormation Frameworks

3

 

HOW SAP NETWEAVER MAKES ANALYTICAL DATA AVAILABLE TO ITS USERS?

In S SA AP BI (as well as in fforme ormer SAP BW), you design Cub Cubes es (I (Info nfoCube Cubes) based on ‘‘e end--u user requir irements ts’. ’. Each Cube has one Fact table (usually v very ery large, sto stores res only measures) that iis s linked linked to many D Dimensio imensions ns tables (relativ (relatively ely small compared to tthe he Fact table). When a user queries against a Cube in SA SAP P BI using BEx (o (orr any other other access tool tool), ), the relevant data is fetched for IInf nfoCubes oCubes (physicall (physically y a set of relatio relational nal tables organiz nized ed in a star schema) fro from m database server up in tthe he appli pplica cati tion on server cache in the fform orm of a virtua virtual multi multidimensional dimensional structure (t (to o support navigation) navigation).. Subsequent queries wil illl be served fro from m thi this s cache (at application applicatio n server level) and when further data is needed, the SAP B BII analyti lytic c engine will request the Data Data Manager to fetch mo more re data ffro rom m the database server.  As Cu Cube v vo olume g grrows or when d diimensions c co omplexities in inc crease ((s snow-fl -flaked, timed based, mult ultip iplle hiera hierarchi rchies, es, etc.) etc.),, the data access perfo performa rmance starts to degrade rapidl rapidly. y. When this this happens, yo you u buil build d aggregates in SAP BI based on small user data navig igattiionviie ews – the frequ frequent queries. Next ti time me, when user issues a query, tthe he analytic engine uses these aggregates to fulf ulfililll user requests with without out accessing the llarg arger Cube. T Thi his s speeds up data access because D Data ata Manager has to do llit ittl tle e wor ork k to lloo ook k up data fr from om aggregates in instead stead of the full cubescan. SAP BI supports two directly access-able data stores for reporting reporting and analysis purpose - Cubes (I (Inf nfoCubes) oCubes) and Opera Operatio tiona nal Data DataStore (OD (ODSObjects – now renamed as DSO, DSO, Data DataStore Objects Objects in SAP SAP NetWe NetWeaver 2004s). ODS objects are flat flat database tables without ithout dimensional structures. The The ODS objects objects can contain any type of data, structured, unstructured or blobs, blobs, while Cubes hold hold only only numeric data (measures) iin n its its Fact tables. tables. HOW SAP BI ACCELERATOR WORKS?

Simply said, the BI accelerator iis s an appli pplianc ance – pre-i pre-insta nstall lled ed 64-bit 4-bit IIntel® ntel® Xeon® processor-base processorbased blade servers from HP and IBM, running Li Linux nux Operatio Operation n System. The TRE REX X (t (text ext retrieval and classif classificatio ication) n) engine with iinn-me memory attri ttribute bute search engine fo forr structured data is iinstalled nstalled o on n each blade (say, 4 bl blad ades, 4 server iinstance nstances, 4 T TRE REX X instances) to support pa para rall llel el search capability. bility. It has ffil ile e system fo forr storage (f (for or Indexes/ data) but no database eng ngine. ine.  At present, you establish an RFC connection between one SAP BI iin nstance to one BI accelerator. Then in each SAP BI instance, define remote links links to the BI accelerator insta instance. In other words, an SAP BI accelerator can act as aunif unified ied enterprise wide wide business intelli intellige gence hub by sourcing raw data from several SAP BI BI instances. HOW W DO SAP BI AND BI ACCEL ACCELERATOR ERATOR WORK WORK TO TOGATHE GATHER? R? HO

In an SAP NetWeaver 2004s BI instance you defi define ne an RFC li link nk ffor or the BI accelerator instance. instanc e. When you ‘RO ‘ROLL LLUP’ UP’ a c cube ubes in SAP BI BI,, you elect the Cub Cube e to be roll rolled-up ed-up (aggregates) iin n the conn connected ected BI accelerator tor inst instanc ance. Cube D Dimensio imensions ns areevenly sto stored red on individual blade individual s. whe Onreby the other ha hand, nd, Faa ct ta table ble isindivi hori horizo zonta ntall y(v pa rtiti rtitioned oned and stored a cross all blade blades, blad columns cathe n be cce ssed individua dua ll lly ylly (vertical ertical decomposi ompositi tion) on)   Copyright Copyright ©2006, Inf nformation ormation Frameworks

4

 

Both, indexes and data is moved to the BI accelerator Both, tor fi file le system. This helps the BI accelerato torr server on each bl blade adeto not onl only y search the iindexes ndexes but also the associated data as well at li lightening ghtening speed – all iin n parallel using all blades. When a user issues a BE BEx x quer query y or access throug through h BA BAPI PI,, o orr A API PI ba based sed appli pplica cati tion, on, the analyti lytic c engine (earli lier er BW O OLA LAP P Processor) kno knows ws that a BI accelerator iis s attached to the SAP BI instance. T The he analytic engine simply routes the query to the BI accelerator by bypassing the relatio relational nal store (skip (skipping ping aggregates or Cube tables in SAP BI BI). ). T The he analytic lytic engine ine fo forma rmats th the e user query in a BI accelerato leratorr-prop proprietary rietary language and passes it alo along ng to the BI accelerator where BI BIA A servers running on each blade runs at ffull ull throt throttl tle e to search what user has requested and hands ov over er cont content ent back to analyt lytic ic engine which int interna ernall lly y maps back to user requested fo forma rmat. The relatio relational nal database is co completely mpletely bypassed in thi this s situation situa tion (Fact/ Dimension Tab Tables). les). Originall riginally, y, T TREX REX search engine emplo ployed yed techniques such as K KNN NN,, Centroi Centroid d and SVM classification on top of a vector space model for classification of content (used for unstructured content – documents; SAP N NetWea etWeaver Po Portal rtal uses these search algori lgorithms) thms) but for BI accelerato tor, r, SA SAP P has now iimplem mplemented an iinn-me memory attri ttribute bute search engine for for structured data that does ad hoc aggregati tion on at a li lighte ghtenin ning g speed. In ot other her words, you bui build ld aggregates on tthe he ffly ly in BI accelerator based o on n user query. T This his is the reason why the very fits irst time ti me aquery is so some me w hat sl slower ower beca use BI acc build d the aggregate in me mory, li like ke OL OLA AP Ca Cach che e in SAP BI Applicatio pplication n elerator Server erver. needs to buil The BI accelerator tor llit itera erall lly y acts a as s an inin-mem memory databa base se fo forr complete SAP BI Cubes providing provi ding access through the BIA engine instead of SQL. In SAP NetWeaver vocabu vocabulary, lary, BI accelerator serv serves es as a cache at the the SAP BI appli applicatio cation n server and is at the same ti time me a full Cube / Aggregate.  Also notice here that SAP BI supports remote Info  Als InfoC Cubes as well as Multi-Pr -Providers (union of mult ltipl iple e physical IInf nfoCubes). oCubes). A large number of SAP BI customers have implemented Multi ulti--Prov Providers. iders. Not Notice ice that, to toda day BI ac accele celerator does not support M Multiulti-Provi Provide ders, but I assume assu me one can simulate multi ulti--prov provider ider li like ke model ((unio union n of Cubes) programmaticall lly y throug thro ugh A API PIs s against underlying aggregated IInf nfoCubes oCubes in B BII Accelerator tor (and much ffaste aster). Note here that at present, you c Note can an not use BI accelerator fo forr OD ODS S tables. It works only only for for Cubes defined in SAP BI BI.. Perhaps future versions of BI accelerator will incorporate other SAP NetWeaver data stores as well. ell. H However, owever, toda today, you have to have Cubes in SAP BI before befo re you ca can n use BI accelerator. Period.  A  ANA NALYS LYSIS

The BI accelerator iis s simply an appliance for SAP BI. You have to have SAP BI BI.. In SAP BI,, yo BI you u model and construct Cubes – Meaning, you have to get user requirements. Y You ou have to define Facts. Y You ou have to define Di Dime mensions. You have to define complete Data Warehouse A Architecture rchitecture – starting from User Requirements – E Extractio xtraction n – Data Transformation – Data Mapping – Data Quali lity ty and Di Distributi stribution on and Data Access Model and datalo load ad strategy in Cubes befo before re you touch BI accelerator. tor.

  Copyright Copyright ©2006, Inf nformation ormation Frameworks

5

 

This his lac lack of clari arity ty rre esult ulte ed in in confusiion on among SA SAP P Customers when Shai emphasiz ized repeatedly ‘‘N No  MoreC eCu ubes’. To use BI accelerato tor, r, to toda day, yo you u have to have Cubes in SAP B BII. When Shai emphasized ized twice in his keynote ‘No‘No-Moreore-Cubes’ Cubes’ is not what it means rather a very fo forw rward-looki looking ng statement. With BI accelerato torr in place, eli eliminate minates the need fo forr aggregates in SAP BI BI.. A Also lso,, data modeling may be si simplif mplified, ied, as many restraining perf performanc ormance consi considera derati tions ons are no longer valid. Mov oving ing fforwa orward, SAP intends to make Inf InfoCubes oCubes optio optiona nall in the process, and then indeed it wil illl conceptu conceptua ally lly be perf rfec ectly tly ffine ine to have ‘no more Cubes’. O Off course, that does not mean that business business process- or data model odeling ing won’t be required anymore. Data models can be kept more generic, as the data modeler will no llonger onger required to extensivel extensively y ‘pre‘pre-think’ think’ (o (orr anticipate) future requirements but such broader general purpose model and lack of ‘endusers prepre-thi thinking’ nking’ will add more complexity iin n secured access proto protocol cols s especiall cially y when inf informatio ormation n has to be aggregated ffrom rom several llay ayers supporti supporting ng dif diffferent generic models. Such approach raises several priv privac acy and security issues (t (typical ypical iissues ssues wit ith h search o oriented riented models). In nutshell, BI accelerator tor may simply a few areas in iimplem mplementing SAP BI but at the same time wil willl o open pen new challenges on tthe he designdesign-runrun-ma managementement-ti time me practices and establi blished shed business intel intelli lige gence best practices and archit rchitecture ectures.  And to defi  And fin ne a cube (to be physical instance or an aggregated cube in BIA IA)) you must also have to gather user business and access requirements an  and d think of all possibl possible e ways to archit chitect ect a high performance Business IIntell ntelligence igence sol solutio ution. n. When making avail vailable able all sort of corpo corporate rate inf informatio ormation n wit ithout hout endend-user user access and content content requirements, yo you u wil illl endend-up up wit ith h an enviro environme nment like the Internet. T Tons ons of qu quickly ickly ac acce cessible inf informa ormatio tion n but not relevant what the user is lo looki oking ng for – no cont context. ext. So business access requirements have to be at the hart of any data/ inf informatio ormation n access strategy to architect an vi viable able ecosystem best ali aligned gned wit ith h the business and their endend-user user needs – bot both h structured &unstructured content. For Da Data W Wa areh rehouse ouse pr pros, os, tthe he concept of BI accelerator iis s similar to g goo ood d old HO HOLA LAP, P, althoug altho ugh the techno hnolo log gy and approach is radicall dically y dif diffferent. Meanin ning g, the content content is transfo nsform rmed ed into pro proprieta prietary structures in another layer on top o off Relatio tiona nall-O OLA LAP P implementa implem enta tion. tiotio n.n User lla ayetio r na sends incoming queries toence HO HOhe LA LAP orthaquick a ccess/ na viga tion insteaac d ces ofs Rela tiona l-OLA LAP. P. T The he only dif diffferen er ce reP isffor t BI accelerato torr uses po pow werful search engine ttechnolo echnology gy, transparent to traditio traditiona nal data wareho rehouse use end users. entioned ned in the keynote) iis s quite appeali ling ng to SAP SAP BI T he B I acce ccele lerat rator or cost   (as mentio customers. I see more applicatio applications ns of BI accelerator in other other intense data access environments. envi ronments. Fo Forr example, the master data management iimplem mplementation is o only nly benef benefici icial al when missi mission on criti critica cal applicatio applications ns use acentral master data management inf infra rastructure. I see next SAP NetWeaver MDM impl implem ementatio entation n to use BI accelerator tor li like ke appli appliance ance ffor or high performance master data loo lookups kups. I can thi think nk of several innov innova ativ tive e implementatio tion no off BI  Acc  Ac celerator. I w wiill d diiscuss a few such scenarios of B BII Ac Acc celerator in my upcoming book ‘Analyti ‘Analytics cs D Design esign Strategies’ to be publi published shed late September 2006.

  Copyright Copyright ©2006, Inf nformation ormation Frameworks

6

 

When looking looking at the BI accelerator tor prici pricing, ng, it takes me back to Shai’ i’s s keynote at the SAPPHI PPHIRE. RE. Sha haii stated stated.. “… One Oneof of their customers was consiide deri ring ngtw two opti ption ons to buil build a hi hig gh perform rformancedata access envir nviron onment. Thetw hetwo opti tion ons plan to accelerate ratethe their Data Warehouse ePr Prog ogram ram– – One, the theH Highly SMP soluti solution on with man many CP CPUs Us on vary very llarg arge box with a databasefrom from… …a databasevendor… [andanother opti ptio onthe theSA SAP P BI accelerator] rator].. We thof put compariis son on what happened between th the ettwo optiio ons. 3 M i ll llii ons Euros worth cense on the theD Data Warehouse from  Ha  H ardwarewith ½ a M i ll llii on E uro uross  worth of lilic them. Comparre ed to 150 T ho lades based IIB BM box, Intel based housa usands nds E uros  bla processor with many many processors but sttiill only 1 15 50 Thousands E Eu uros [[B BI accelerato tor] option. Nowthe theiinteresting stingthing thingiis not jus just howmuch it costs us to acquirethe retheha hardware and softw tware, it is is how much iitt costs you to manage and operate te these ettwo enviirronments [One Oneb being ngD DatabaseV endor Optio tion vs. SA SAP P BI accelerator]. r]. If you look howmany  DB  D BAs it needs to maintain and operateo eon n stored data warehouseof this magnitudevs evs. what you need todowith the thebo box that lliiteral rallly plugs iin. n. W Whi hirl rlpo pool pl plug ugged in this this box [BI [BI accelerator] within thin one week and lilive veand and producti tive veiin one week. That what you get in  No omoreD eDB B, noMoreAdmin, NomoreC eCo ost…” compari riso son... n.... ---- N Shai A gassi, Keynoteat teat Sahhpire’06, Orl Orlan ando, May17, 2006.

Humm … are we comparing Apples pples and Orang Oranges here? We are talking lking about an Appl Appliance iance vs. an Entire Ent ire Data Data Warehouse Infra Infrastructure. It It is correct correct to say that BI accelerator tor is an appliance appliance that Turbo Charges SAP BI data access and that’s allll.. In other words we are comparing cost of of “ “high high octane gas’ o orr ‘Turbo ‘Turbo Charger’ for for a Car to perform perform better vs. a brand new high-end h-end race CAR. Not a right comparison. The best comparison would be to add SAP BI costs (HW/ SW/ Support/ DBA BA)) plus BI accelerato torr and then compared with the the new option option for for a Data Data Warehouse Solutio olution n ffrom rom another vendor(s). Tha That would have been aright comparison. Theretoo, SAP soluti solution on will relativ relatively ely win but this this will ill be a right way to justif justify y and compare.  CONCLUSIONS

In my opinion, SAP BI with BI accelerator option will be a win-win proposition for highend SAP BI customers, no nott necessary fro from m HW/ SW/ Support co costs sts angles but from what business value SAP BI will add naturally to your SAP iinvestment nvestment such as prepre-de defined business content content (t (the he most compl complex ex and time consuming task), analyti lytics, cs, meta-model and continue conti nued ffuture uturebuilt built--in capabiliti bilities es that will come n na aturall lly yw with ith SAP BI as SAP adds other business sol solutio utions ns a and nd most important value of SAP BI being an integral part of NetWeaver to support its ESA driven intelligence business solution. I stro strongly ngly agree that the BI accelerator tor is a great op opti tion on ffor or SAP BI customer wit ith h large SAP BI data warehouses. Ev Even en wit ith h BI accelerator, tor, SAP B BII customers need to remember that they need to be very careful iin n archit rchitecting ecting their enterprise business int intell elligence igencesol solutio ution n that is aligned wit ith h emerging business and technol technology ogy archit hitectures ectures (Servi ervice–O ce–Oriented) riented) which requires embedded-analytics rather typi typica cal reporti reporting ng and analysi lysis s model odel.. The key is to llist isten en   Copyright Copyright ©2006, Inf nformation ormation Frameworks

7

 

to user’s needs. How they want to use inf informa ormati tion on tto o make right decisio decisions. ns. A And nd remember; only onl y a handf ndful ul o off users wil illl require detailed data but most will be just as happy to get minimum data points tto o meet their needs. And iiff you plan to use BI accelerator, you sti still ll have to maintain your SAP BI infra infrastruc structure ture – Cub Cubes es (f (for or now), O OD DS implementatio tions, ns, ETL, ET L, Data Lo Load ads, Maint intena enance etc. And most important ffac act to remember is that you need to be quite careful how you model o objects bjects that to meet your needs without BI accelerator and w with ith BI accelera lerator. tor. Otherwise, you wil willl endend-up up redesigning c core ore SAP BI obj objec ects ts when  y  yo ou pl plan to to implement BI accelerator. So you ha have to to think o off possibility o off BI accelerator implementation tion at the design time of SAP BI objects objects. In nutshell, when Shai state stated N  No ‘ o MoreCu eCubes’, perhaps he was referring to ‘no more aggregates in SAP BI’ and for y ‘ ou do not need to pre-think of any question that any user will ask’, perhaps, he meant ‘no user runn-time time ad hoc queries definiti efinition’. on’.

 Ab bout The Author: Naeem Hashmi, Chief Research Off Offic icer, er, I Inform nformation F Fram rameworks  A

  Naeem is a worldworld-renowned renowned thought llea eader, author, speaker, and expert on on emerging Inf nforma ormatio tion n T Technologies echnologies with 25+ years of experience in Enterprise Archit Architecture ectures, Distributed Service-O ervice-Oriented Business applicatio applications ns design, ERP application application Integ Integration, Inf nforma ormatio tion n D Deliv elivery ery Archit Architecture ectures, D Data ata Warehousing; Data D ata Mining; ining; CRM, Management Consulting, Consulti ng, Knowledg Knowledge and Technology Technology T Tran ransfer and visualizing visualizing new products. Current research work includes Inf Informa ormation tion Vi Visua sualization, lization, Social Networks, IInfo nforma rmation tion Psyche Mining, and the Enterprise Architectures. Architectures. Naeem is Founder, Chief Chief Research Officer fo forr Inf Information ormation Frameworks Naeem advises IT vendors on innovative technologie technologies s and produc productt He authored/ coco -authored four boo books; ks; the best selling book ‘Business  Information Warehouse for SAP AP’’, a contributing author for upcoming book ‘Technologies  f  fo or Government Transformati ation on: ERP Systems and Beyond’ edited by Shayne Kavanagh et.al. To To be published published by the Gov overnm ernment Finance Officers Association. Naeem is a frequent speaker at national and international IT conferences. He has authored and published well over over 100+ articles on on IInf nformation ormation T Technolo echnologies gies and has authored several cover cover stories stori es and feature for the the Intelli Intellige gent Enterprise Enterprise Magazine, IIntelli ntellige gent ERP ERP Magazine, Manufacturing Systems International tional and the CIO magazine. zine. Naeem served a disti distingu nguished judge for Intelli I ntellige gent E Enterprise nterprise Award for for IT Excellent 2005 in Oct 2005. Upcoming Books Books::  An  Ana seArc Desicg s.dTo beep un bc lieM shed 20b0ep 6 ublished. Ja  Entaelryptric is eAr hnitS ecttruarteesgiaen Gov rna o:dS ele sp . t.To Jan n. 2007 Naeem can b be e reached at [email protected] [email protected]   T: 603-432-4550 M: 603-66 603-6611-682 6820 Web We b Site http:/ / infof infofra rameworks.com

 AC CKNO NOW WLED LEDG GMENTS  A  Author wouldliketoacknowledgesupport of Business IntelligenceAccelerator Teamat SAP for briefingtheir solution andJens Derpmund, Senior Solutio ionArchitte ect ffo or SAP AP Ne NetWe Weaver BI Teama mat HP HP fo for reviewinga gandprovid iding input for this is paper.

D iscla isclaime imer: r:    As SAP NetWeaver Business IntelligenceAccelerator isstill evolving, thispaper analysisisbasedonbest information th

availa lablef efor thep epresent and near stateo eof Business Intelli ligenceAccelerator as of May 19 , 2006.   Copyright Copyright ©2006, Inf nformation ormation Frameworks

8

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close