Introduction to Ab Initio Prepared By : Ashok Chanda
Ab initio Session 1
Introduction to DWH Introduction D WH Explanation of DW Architectur Architecture e Operating System / Hardwar Hardware e Support Introduction to ET !rocess Introduction Introductio n to A" Initio Explanation of A" Initio Architecture Architecture
What is Data Warehouse
A data warehouse is a copy of transaction data speci#cally structured for querying and reporting. A data warehouse is a su"$ect%oriented& integrated& time%'ariant and non%'olatile collection of data in support of management(s decision ma)ing process* A data warehouse is a central repository for all or signi#cant parts of the data that an enterprise(s 'arious "usiness systems collect*
Data Warehouse-Defnitions
A data warehouse is a data"ase geared towards the "usiness intelligence re+uirements of an organi,ation* The data warehouse integrates data from the 'arious operational systems and is typically loaded from these systems at regular inter'als* Data warehouses contain historical information that ena"les analysis of "usiness performance o'er time* A collection of data"ases com"ined with a -exi"le data extraction system*
Data Warehouse
A data warehouse can "e normali,ed or denormali,ed* It can "e a relational data"ase& multidimensional data"ase& -at #le& hierarchical data"ase& o"$ect data"ase& etc* Data warehouse data often gets changed* And data warehouses often focus on a speci#c acti'ity or entity*
Why Use a Data Warehouse
Data Exploration and Disco'ery Integrated and .onsistent data uality assured data Easily accessi"le data !roduction and performance awareness Access to data in a timely manner
Simplified Datawarehouse Architecture
Data !arehouse Ar"hite"ture
Data Warehouses can "e architected in many di0erent ways& depending on the speci#c needs of a "usiness* The model shown "elow is the 2hu"%and%spo)es2 Data Warehousing architecture that is popular in many organi,ations* In short& data is mo'ed from data"ases used in operational systems into a data warehouse staging area& then into a data warehouse and #nally into a set of conformed data marts* Data is copied from one data"ase to another using a technology called ET 3Extract& Transform& oad4*
#he $#% Pro"ess
.apture Scru" or Data cleansing Transform oad and Index
$#% #e"hno&ogy
ET Technology is an important component of the Data Warehousing Architecture* It is used to copy data from Operational Applications to the Data Warehouse Staging Area& from the DW Staging Area into the Data Warehouse and #nally from the Data Warehouse into a set of conformed Data 5arts that are accessi"le "y decision ma)ers* The ET software extracts data& transforms 'alues of inconsistent data& cleanses 2"ad2 data& #lters data and loads data into a target data"ase* The scheduling of ET $o"s is critical* Should there "e a failure in one ET $o"& the remaining ET $o"s must respond appropriately*
Data Warehouse Staging Area
The Data Warehouse Staging Area is temporary location where data from source systems is copied* A staging area is mainly re+uired in a Data Warehousing Architecture for timing reasons* In short& all re+uired data must "e a'aila"le "efore data can "e integrated into the Data Warehouse* Due to 'arying "usiness cycles& data processing cycles& hardware and networ) resource limitations and geographical factors& it is not feasi"le to extract all the data from all Operational data"ases at exactly the same time
$'a(p&es- Staging Staging Area
6or example& it might "e reasona"le to extract sales data on a daily "asis& howe'er& daily extracts might not "e suita"le for #nancial data that re+uires a month%end reconciliation process* Similarly& it might "e feasi"le to extract 2customer2 data from a data"ase in Singapore at noon eastern standard time& "ut this would not "e feasi"le for 2customer2 data in a .hicago data"ase*
Data in the Data Warehouse can "e either persistent 3i*e* remains around for a long period4 or transient 3i*e* only remains around temporarily4*
7ot all "usiness re+uire a Data Warehouse Staging Area* 6or many "usinesses it is feasi"le to use ET to copy data directly from operational data"ases into the Data Warehouse*
Data !arehouse
The purpose of the Data Warehouse in the o'erall Data Warehousing Architecture is to integrate corporate data* It contains the 2single 'ersion of truth2 for the organi,ation that has "een carefully constructed from data stored in disparate internal and external operational data"ases* The amount of data in the Data Warehouse is massi'e* Data is stored at a 'ery granular le'el of detail* 6or example& e'ery 2sale2 that has e'er occurred in the organi,ation is recorded and related to dimensions of interest* This allows data to "e sliced and diced& summed and grouped in unimagina"le ways*
Data Warehouse
.ontrary to popular opinion& the Data Warehouses does not contain all the data in the organi,ation* It(s purpose is to pro'ide )ey "usiness metrics that are needed "y the organi,ation for strategic and tactical decision ma)ing* Decision ma)ers don(t access the Data Warehouse directly* This is done through 'arious front%end Data Warehouse Tools that read data from su"$ect speci#c Data 5arts* The Data Warehouse can "e either 2relational2 or 2dimensional2* This depends on how the "usiness intends to use the information*
Data Warehouse $n)iron(ent In addition to a relational/multidimensional data"ase& a data warehouse en'ironment often consists of an ET solution& an OA! engine& client analysis tools& and other applications that manage the process of gathering data and deli'ering it to "usiness users*
Data *art
A su"set of a data warehouse& for use "y a single department or function* A repository of data gathered from operational data and other sources that is designed to ser'e a particular community of )nowledge wor)ers* A su"set of the information contained in a data warehouse* Data marts ha'e the same de#nition de#nition as the data warehouse 3see "elow4& "ut data marts ha'e a more limited audience and/or data content*
Data *art
ET 3Extract Transform oad4 $o"s extract data from the Data Warehouse and populate one or more Data 5arts for use "y groups of decision ma)ers in the organi,ations* The Data 5arts can "e Dimensional 3Star Schemas4 or or relational& depending on how the information is to "e used and what 2front end2 Data Warehousing Tools will "e used to present the information* Each Data 5art can contain di0erent com"inations of ta"les& columns and rows from the Enterprise Data Warehouse* 6or example& an "usiness unit or user group that doesn(t re+uire a lot of historical data might only need transactions from the current calendar year in the data"ase* The !ersonnel Department might need to see all details a"out employees& whereas data such as 2salary2 or 2home address2 might not "e appropriate for a Data 5art that focuses on Sales*
Star S"he(a
The star s"he(a is is perhaps the simplest data warehouse schema* It is called a star schema "ecause the entity%relationship diagram of this schema resem"les a star& with points radiating from a central ta"le* The center of the star consists of a large fact ta"le and the points of the star are the dimension ta"les*
Star S"he(a + "ontinued
A star schema is characteri,ed "y one or more 'ery large ,a"t ta"les ta"les that contain the primary information in the data warehouse& and a num"er of much smaller di(ension ta"les ta"les 3or loo)up ta"les4& each of which contains information a"out the entries for a particular attri"ute in the fact ta"le*
Ad)antages o, Star S"he(as
!ro'ide a direct and intuiti'e mapping "etween the "usiness entities "eing analy,ed "y end users and the schema design* !ro'ide highly optimi,ed performance for typical star +ueries* Are widely supported "y a large num"er of "usiness intelligence tools& which may anticipate or e'en re+uire that the data% warehouse schema contain dimension ta"les Star schemas are used for "oth simple data marts and 'ery large data warehouses*
Star s"he(a
Diagrammatic representation of star schema
Sno!ake S"he(a
The snow-a)e schema is a more complex data warehouse model than a star schema& and is a type of star schema* It is called a snow-a)e schema "ecause the diagram of the schema resem"les a snow-a)e* Snow-a)e schemas normali,e dimensions to eliminate redundancy*
Sno!ake S"he(a $'a(p&e
That is& the dimension data has "een grouped into multiple ta"les instead of one large ta"le* 6or example& a product dimension ta"le in a star schema might "e normali,ed into a products ta"le& a product8category ta"le& and a product8manufacturer ta"le in a snow-a)e schema* While this sa'es space& it increases the num"er of dimension ta"les and re+uires more foreign )ey $oins* The result is more complex +ueries and reduced +uery performance*
Diagra((ati" representation ,or Sno!ake S"he(a
a"t #ab&e The The centrali,ed ta"le in a star schema is called as 6A.T ta"le* A fact ta"le typically has two types of columns9 those that contain facts and those that are foreign )eys to dimension ta"les* The primary )ey of a fact ta"le is usually a composite )ey that is made up of all of its foreign )eys*
What happens during the $#% pro"ess
During extraction& the desired data is identi#ed and extracted from many di0erent sources& including data"ase systems and applications* Depending on the source system(s capa"ilities 3for example& operating system resources4& some transformations may ta)e place during this extraction process* The si,e of the extracted data 'aries from hundreds of )ilo"ytes up to giga"ytes& depending on the source system and the "usiness situation* After extracting data& it has to "e physically transported to the target system or an intermediate system for further processing*
$'a(p&es o, Se"ond/eneration $#% #oo&s
Po!er(art 0. + 2n,or(ati"a Corporation
Ardent DataStage + Ardent So,t!are3 2n".
!rogressi'ely integrated with 5icrosoft
Ab 2nitio 6.6 + Ab 2nitio So,t!are
:eneral%purpose tool oriented to data marts
Sagent Data *art So&ution 4.5 + Sagent #e"hno&ogy
!ioneer due to mar)et share
A )it of tools that can "e used to "uild applications
#apestry 6.1 + D673 2n"
End%to%end data warehousing solution from a single 'endor
What to &ook ,or in $#% too&s
Use optiona& data "&eansing too& to "&ean-up sour"e data Use e'tra"tion8trans,or(ation8&oad too& to retrie)e3 "&eanse3 trans,or(3 su((ari9e3 aggregate3 and &oad data Use (odern3 engine-dri)en te"hno&ogy ,or ,ast3 para&&e& operation /oa&: defne 155 o, the trans,or( ru&e !ith point and "&i"k inter,a"e Support de)e&op(ent o, &ogi"a& and physi"a& data (ode&s /enerate and (anage "entra& (etadata repository ;pen (etadata e'"hange ar"hite"ture to integrate "entra& (etadata !ith &o"a& (etadata. Support (etadata standards Pro)ide end users a""ess to (etadata in business ter(s
;perating Syste( 8
This section discusses how a D;5S utili,es OS/hardware features such as parallel functionality& S5!/5!! support& and clustering* These OS/hardware features greatly extend the scala"ility and impro'e performance* Howe'er& managing an en'ironment with these features is di
Para&&e& un"tiona&ity
The introduction and maturation of parallel processing en'ironments are )ey ena"lers of increasing data"ase si,es& as well as pro'iding accepta"le response times for storing& retrie'ing& and administrating data* D;5S 'endors are continually "ringing products to mar)et that ta)e ad'antage of multi%processor hardware platforms* These products can perform ta"le scans& "ac)ups& loads& and +ueries in parallel*
Para&&e& eatures
An o'er'iew of typical parallel functionality is gi'en "elow "elow 9 ueries = !arallel +ueries can enhance scala"ility for many +uery operations Data load = !erformance is always a serious issue when loading large data"ases* 5eeting response time re+uirements is the o'erriding factor for determining the "est load method and should "e a )ey part of a performance "enchmar) .reate ta"le as select = This feature ma)es it possi"le to create aggregated ta"les in parallel Index creation = !arallel index creation exploits the "ene#ts of parallel hardware "y distri"uting the wor)load generated "y a large index created for a large num"er of processors *
Whi"h para&&e& pro"essor "onfguration3 S*P or *PP
S5! and clustered S5! en'ironments & ha'e the -exi"ility and a"ility to scale in small increments* S5! en'ironments are often useful for the large& "ut static data warehouse& where the data cannot "e easily partitioned& due to the unpredicta"le nature of how the data is $oined o'er multiple ta"les for complex searches and ad%hoc +ueries*
Whi"h para&&e& pro"essor "onfguration3 S*P or *PP
5!! wor)s well in en'ironments where growth is potentially unlimited& access patterns to the data"ase are predicta"le& and the data can "e easily partitioned across di0erent 5!! nodes with minimal data accesses crossing "etween them* This often occurs in large OT! en'ironments& where transactions are generally small and predicta"le& as opposed to decision support and data warehouse en'ironments& where multiple ta"les can "e $oined in unpredicta"le ways* In fact& data warehousing and decision support are the areas most 'endors of parallel hardware platforms and D;5Ss are targeting* 5!! does not scale well if hea'y data warehouse data"ase accesses must cross 5!! nodes& causing I/O "ottlenec)s o'er the 5!! interconnect& or if multiple 5!! nodes are continually loc)ed for concurrent record updates*
A *u&ti-CPU Co(puter =S*P>
et!or o *u u t --CPU A ?et!or ?odes
A ?et!ork o, ?et!orks
Para&&e& Co(puter Ar"hite"ture
.omputers come in many >shapes and si,es?9 Single%.!@& 5ulti%.!@ 7etwor) of single%.!@ computers 7etwor) of multi%.!@ computers
5ulti%.!@ machines are often called S5!s 3for Symmetric 5ulti !r !rocessors4* ocessors4*
Specially%"uilt networ)s of machines are often called 5!!s 3for 5assi'ely !arallel !rocessors4* !r ocessors4* !roc essors4*
2ntrodu"tion to Ab 2nitio
Ab 2nitio So,t!are Corporation was was founded in the mid 1@@5s "y "y Sheryl Handler& the former .EO at Thin)ing 5achines .orporation& after T5. #led for "an)ruptcy* In addition to Handler& other former T5. people in'ol'ed in the founding of A" Initio included .li0 asser& Angela ordi& and .raig Stan#ll* A" Initio is )nown for "eing 'ery secreti'e in the way that they run their "usiness& "ut their software is widely regarded as top notch*
The A" Initio software is a fourth generation data analysis& "atch processing& data manipulation graphical user interface 3:@I4% "ased parallel processing tool that is used mainly to extract& transform and load data* The A" Initio software is a suite of products that together pro'ides platform for ro"ust data processing applications* The .ore A" Initio !roducts are9 The B.oCOperating System The .omponent i"rary The :raphical De'elopment En'ironment
What Does Ab 2nitio *ean *ean
A" Initio is atin for >6rom the ;eginning*?
6rom the "eginning our software was designed to support a complete range of "usiness applications& from simple to the most complex* .rucial capa"ilities li)e parallelism and chec)pointing cant "e added after the fact*
The :raphical De'elopment En'ironment and a powerful set of components allow our customers to get 'alua"le results from the "eginning*
Ab 2nitios ,o"us
>5o'ing Data?
High !erformance
mo'e small and large 'olumes of data in an e
;etter producti'ity
Ab 2nitios So,t!are
A" Initio software is a general% purpose data processing platform for mission%critical applications such as9
Data warehousing ;atch processing .lic)%stream analysis Data mo'ement Data transformation
App&i"ations o, Ab 2nitio So,t!are
!rocessing $ust a"out any form and 'olume of data*
!arallel sort/merge processing*
Data transformation*
ehosting of corporate data*
!arallel execution of existing applications*
Ab 2nitio Pro)ides or:
Distri"ution % a platform for applications to execute across a collection of processors within the con#nes of a single machine or across multiple multiple machines*
educed un Time .omplexity % the a"ility for applications to run in parallel on any com"ination of computers where the A" Initio .oCOperating System is installed from a single point of control*
pp "at ons o A 2n nto App So,t!are in ter(s o, Data Warehouse
6ront end of Data Warehouse9
Transformation of disparate sources Aggregation and other preprocessing eferential integrity chec)ing Data"ase loading
;ac) end of Data Warehouse9
Extraction for external processing Aggregation and loading of Data 5arts
Ab 2nitio or 2n,or(ati"aPo!er,u& $#%
Informatica and A" Initio "oth support parallelism* ;ut Informatica supports only one type of parallelism "ut the A" Initio supports three types of parallelism* In Informatica the de'eloper need to do some partitions in ser'er manager "y using that you can achie'e parallelism concepts* ;ut in A" Initio the tool it self ta)e care of parallelism we ha'e three types of parallelisms in A" Initio F* .omponent G* Data !arallelism * !ipe ine parallelism this is the di0erence in parallelism concepts* G* We don(t ha'e scheduler in A" Initio li)e Informatica you need to schedule through script or u need to run manually* * A" Initio supports di0erent types of text #les means you can read same #le with di0erent structures that is not possi"le in Informatica& and also A" Initio is more user friendly than Informatica so there is a lot of di0erences in Informatica and A" initio*
* A"Initio doesn(t need a dedicated administrator& @7IJ or 7T Admin will su
Ab 2nitio or 2n,or(ati"aPo!er,u& $#%-"ontinued
Error Handling % In A" Initio you can attach error and re$ect #les to each transformation and capture and analy,e the message and data separately* Informatica has one huge logK Lery ine
o"ust transformation language % Informatica is 'ery "asic as far as transformations go* While I will not go into a function "y function comparison& it seems that A" Initio was much more ro"ust*
Instant feed"ac) % On execution& A" Initio tells you how many records ha'e "een processed/re$ected/etc* and detailed performance metrics for each component* Informatica has a de"ug mode& "ut it is slow slow and and di
Both too&s are Both ,unda(enta&&y diEerent Which one to use depends on the wor) at hand and existing infrastructure and resources a'aila"le* Informatica is an engine "ased ET tool& the power this tool is in it(s transformation engine and the code that it generates after de'elopment cannot "e seen or modi#ed* A" Initio is a code "ased ET tool& it generates )sh or "at etc* code& which can "e modi#ed to achie'e the goals& if any that cannot "e ta)en care through the ET tool itself* A" Initio doesn(t need a dedicated administrator& @7IJ or 7T Admin will su
Ab 2nitio Produ"t Ar"hite"ture User Userpp'iations pp'iations &eve'opment &eve'opmentnvironments nvironments GDE GDE
#omponent #omponent %i!rary %i!rary
She'' She''
User*de+ined User*de+ined #omponents #omponents
3rd 3rdarty arty #omponents #omponents
! !"nitio "nitio - -
$ The ! "nitio #o >Operating $ The ! "nitio #o>Operating System System
Native NativeOperating OperatingSystem System(Unix, (Unix,Windows, Windows,OS/390) OS/390)
Ab 2nitio Ar"hite"ture$'p&anation
The A" Initio .ooperating system unites the networ) of computing resources%.!@s&storage dis)s & programs & datasets into a production +uality data processing system with scala"le performance and mainframe class relia"ility* The .ooperating system is layered on the top of the nati'e operating systems of the collection of ser'ers *It pro'ides a distri"uted model for process execution& #le management &de"ugging& process monitoring & chec)pointing *A user may perform all these functions from a single point of control*
CoF;perating Syste( Ser)i"es
!arallel and distri"uted application execution
.ontrol Data Transport
Transactional semantics at the application le'el* .hec)pointing* 5onitoring and de"ugging* !arallel #le management* 5etadata%dri'en components*
Ab 2nitio: What We Do
A" Initio software helps you "uild large%scale data processing applications and run them in parallel en'ironments* A" Initio software consists of two main programs9 CoF;perating Syste(: which your system administrator installs on a host @nix or Windows 7T ser'er& as well as on processing computers* #he /raphi"a& De)e&op(ent $n)iron(ent =/D$>: which you install on your !. 3 GDE Computer 4 and con#gure to communicate with the host*
#he Ab 2nitio CoF;peratingG Syste(
#he CoF;perating Syste( uns uns across a 'ariety of Operating Systems and Hardware !latforms including OS/MN on 5ainframe& @nix& and Windows* Supports distri"uted and parallel execution* .an pro'ide scala"ility proportional to the hardware resources pro'ided* Supports platform independent data transport*
#he Ab 2nitio CoF;peratingG Syste(Continued The A" Initio .oCOperating System depends on parallelism to connect 3i*e*& cooperate with4 di'erse data"ases* It extracts& transforms and loads data to and from Teradata and other data sources*
Co-;perating Syste( %ayer Any ;S
:DE
#op %ayer
Solaris& AIJ& 7T& inux& 7.
:DE
Co-;p Syste(
:DE Sa(e Co-;p Co((a ;n any ;S.
:DE
/raphs "an be (o)ed ,ro ;ne ;S to another !8o an Changes.
#he Ab 2nitio CoF;perating Syste( Huns on:
Sun Solaris I;5 AIJ Hewlett%!ac)ard H!% @J Siemens !yramid eliant @7IJ I;5 D7IJ/ptx Silicon :raphics IIJ
ed Hat inux Windows 7T P*N 3xQ4 Windows 7T GNNN 3xQ4 .ompa+ .ompa+ TruQP TruQP @7IJ I;5 OS/MN 7. 5!%AS
Conne"ti)ity to ;ther So,t!are
.ommon& high performance data"ase interfaces9
I;5 D;G& D;G/!E& D;GEEE& @D;& I5S Oracle& Informix J!S&Sy"ase&Teradata&5S S Ser'er R OE%D; OD;.
Other software pac)ages9
.onnectors to many other third party products Trillium& ErWin& Sie"el& etc*
Ab 2nitio Cooperating Syste( A" Initio Software .orporation& head+uartered in exington& 5A& de'elops software solutions that process 'ast amounts of data 3well into the tera"yte range4 in a timely fashion "y employing many 3often hundreds4 of ser'er processors in parallel* 5a$or corporations worldwide use A" Initio software in mission critical& enterprise%wide& data processing systems* Together& Teradata and A" Initio deli'er9 End%to%end solutions for integrating and processing data throughout the enterprise Software that is -exi"le& e
Graphical Development Environment GDE
#he /D$ The :raphical De'elopment En'ironment 3:DE4 pro'ides a graphical user interface into the ser'ices of the .oCOperating System* #he /raphi"a& De)e&op(ent $n)iron(ent Ena"les Ena"les you to create applications "y dragging and dropping .omponents* Allows you to point and clic) operations on executa"le -ow charts* The .oCOperating System can execute these -owcharts directly* :raphical monitoring of running applications allows you to +uantify data 'olumes and execution times& helping spot opportunities for impro'ing performance*
#he /raph *ode&
#he Co(ponent %ibrary:
eusa"le #he Co(ponent %ibrary: eusa"le software 5odules for Sorting& Data Transformation& data"ase oading Etc* The components adapt at runtime to the record formats and "usiness rules controlling their "eha'ior* A" Initio products ha'e helped reduce a pro$ects de'elopment and research time signi#cantly*
Co(ponents
.omponents may run on any computer running the .oCOperating System* Di0erent components do di0erent $o"s* The particular wor) a component accomplishes depends upon its parameter settings* Some parameters are data transformations& that is "usiness rules to "e applied to an input 3s4 to produce a re+uired output*
4rd Party Party Co(ponents
$*$
The Enterprise 5etaCEn'ironment 3E5E4 is a high% performance o"$ect%oriented storage system that in'entories and manages 'arious )inds of information associated with A" Initio applications* It pro'ides storage for all aspects of your data processing system& from design information to operations data* The E5E also pro'ides rich store for the applications themsel'es& including data formats and "usiness rules* It acts as hu" for data and de#nitions * Integrated metadata management pro'ides the glo"al and consolidated 'iew of the structure and meaning of applications and data% information that is usually scattered throughout you "usiness *
Benefts o, $*$ The Enterprise 5etaCEn'ironment pro'ides a rich store for applications and all of their associated information including 9 Technical 5etadata%Applications related "usiness rules &record formats and execution statistics ;usiness 5etadata%@ser de#ned documentations of $o" functions &roles and responsi"ilities* 5etadata is data a"out data and is critical to understanding and dri'ing your "usiness process and computational resources *Storing and using metadata is as important to your "usiness as storing and using data*
$*$-Ab 2nitio He&e)an"e
;y integrating technical and "usiness metadata &you can grasp the entirety of your data processing from operational to analytical systems* The E5E is completely integrated en'ironment* The following #gure shows how it #ts in to the high le'el architecture of A" Initio software*
Step!ise e'p&anation o, Ab 2nitio Ar"hite"ture
ou construct your application from the "uilding "loc)s called components& manipulating them through the :raphical De'elopment En'ironment 3:DE4* ou chec) in your applications to the the E5E* E5E* The E5E and :DE uses the underlining underlining functionality functionality of the .oCOperating System to perform many of their tas)s* The .ooperating System units the distri"uted resources into a single > 'irtual computer? to run applications in parallel* A" Initio software runs on @nix &Windows 7T&5LS operating systems*
tep! se e'p anat on o A Step! 2nitio Ar"hite"ture "ontinued
A" Initio connector applications extract metadata from third part metadata sources into the E5E or extract it from the E5E into a third party destination* ou 'iew the results of pro$ect and application dependency analysis through a We" user interface *ou also 'iew and edit your "usiness metadata through a we" user interface*
$*$ :Iarious users "onstituen"y ser)ed The E5E addresses the metadata needs of three di0erent constituencies9 ;usiness @sers De'elopers System Administrators System
$*$ :Iarious users "onstituen"y ser)ed
;usiness users are interested in exploiting data for analysis& in particular with regard to data"ases &ta"les and columns* De'elopers tend to "e oriented towards applications &needing to analy,e the impact of potential program changes* System Administrator and production personnel want $o" status information and run statistics*
$*$ 2nter,a"es
We can create and manage E5E through interfaces9 :DE We" @ser Interface Air @tility
