Download This is the html version of the file http://devnet

Informatica Metadata Exchange Frequently Asked Questions
Q: What is Informatica Metadata Exchange?
A: Informatica Metadata Exchange is a family of metadata exchange products for the
Informatica Data Integration platform. The Informatica Metadata Exchange Software
Development Kit (MX SDK) provides an open, bi-directional application programming interface
(API) for metadata integration between the Informatica Metadata Repository and other
metadata stores.
Informatica Metadata Exchange tools leverage the MX SDK to provide metadata integration
with data models, relational database catalogs, ERP systems and other metadata
sources.
Q: How is Informatica Metadata Exchange different from MX SDK?
A: Metadata Exchange is the umbrella term applied to all of Informatica’s metadata exchange
products. MX SDK is the Metadata Exchange Software Development Kit, the enabling
technology that the other Metadata Exchange products are built on.
Q: Who should use Informatica Metadata Exchange?
A: Metadata Exchange provides tools for Informatica administrators and developers.
Administrators can use the Metadata Exchange products to help them manage their metadata
in a more timely, accurate, and efficient manner. Developers can build applications that use
the MX SDK to deliver metadata to business users.
Q: How comprehensive is the MX SDK?
A: The MX SDK supports most metadata objects in the Informatica repository including
Repositories, Folders, Sources, Targets, Transformations, Mappings, Mapplets, Schedules,
Workflows, Workflow logs, Worklets, Sessions, Database Connections, Metadata Extensions
and many others.
Q: Can I use MX SDK to update sources and targets?
A: Yes. The MX SDK source and target objects are fully updateable. You can rename the
object, change business names and descriptions, add, modify or remove columns, update
indexes (targets only) and foreign key relationships and update metadata extensions. The only
limitation is that MX SDK does not allow you to change the database platform for an existing
source or target – in this case you must delete the object and recreate it.
Q: Can I use MX SDK to update transformations?
A: Not all transformations can be updated using the MX SDK. MX SDK support for
transformations allows developers to update the properties of their external procedure
transformations and advanced external procedure transformations. Standard transformations
are read-only in MX SDK and have to be updated using Designer.
Q: Can I use MX SDK to update mappings?
A: No. Because of the potential complexity of an Informatica mapping, the MX SDK mapping
objects are read only. Mappings must be created using Informatica Designer.
Q: Can I use MX SDK to find out when a workflow was run and if it succeeded?
A: Yes. The MX SDK provides a rich object model for querying workflow, session task and
transformation logs.
1
Q: Can I use MX SDK to determine data lineage?
A: MX SDK can determine data lineage at a source or target level, but not at a field/column
level. The source and target objects have methods for determining shortcut, source, target
and mapping dependencies.
The dependency information returned by these methods is the same as the
dependency information that is visible in the Informatica Repository Manager.
More detailed information about dependencies within a mapping is available when you use MX
SDK or Designer to export the mapping as an XML file.
Q: Can I use Informatica Metadata Exchange to manage my own custom metadata?
A: Informatica metadata extensions were introduced as a new feature in Informatica
PowerCenter 6.0. You can add your own custom metadata extensions to sources, targets,
mappings, transformations, workflows and other objects using the Informatica Repository
Manager.
These metadata extensions then become part of the repository metamodel and can
be maintained using the Informatica Designer and viewed by end-users in the Informatica
Metadata Reporter. These metadata extensions can also be read and updated by other
applications that use the MX SDK. This means that Metadata Exchange for Data Models can be
used to populate source and target metadata extensions from data models, database catalogs
and other Informatica repositories.
Q: I am an Informatica Developer. Which Informatica Metadata Exchange tool is
right for me?
A: You need Metadata Exchange for Data Models. You can design and document your
database schema using a data modeling tool, then use Metadata Exchange for Data Models to
replicate your data model in the Informatica repository, creating either source or target
definitions. As you refine your data models, your database schemas and your mappings,
Metadata Exchange for Data Models helps you to detect and resolve metadata
discrepancies that may cause your mappings to fail. The tool’s validation, change and
dependency reports can help to identify the impact of schema changes. Used in conjunction
with Informatica metadata extensions, you can document your design with rich metadata,
including business and technical definitions, metadata and data lineage, update timestamps,
volumetric data and extended metadata from your data models.
Q: I am responsible for administering a data warehouse. Which Informatica
Metadata Exchange tool is right for me?
A: Metadata Exchange for Data Models will help to ensure that your source and target
definitions are always consistent with the databases that they represent. If the database
schema changes, validation, change and dependency reports will identify the impact of the
change. Metadata Exchange for Data Models will also help to ensure that business and
technical metadata is always current.
If your data warehouse contains data that originates in a SAP or Peoplesoft ERP system,
PowerConnect Metadata Exchange gives you tools for automated preventative maintenance
and will ease the pain of managing an ERP upgrade or customization. This product
dramatically reduces the work required to keep the Informatica repository current with
changes in an ERP system by providing tools to manage ERP metadata. You can quickly and
selectively identify metadata differences between the ERP system and the Informatica
repository, determine the impact of the discrepancy and optionally update the Informatica
repository to resolve the problem.
2
Q: I need to deliver business metadata to my business users. Which Informatica
Metadata Exchange tool is right for me?
A: Informatica provides the Informatica Metadata Reporter, a simple web-based reporting
solution. If this does not fit your needs, many Business Intelligence tool vendors have prebuilt integration tools for delivering Informatica metadata to end-users (for more information
see the Informatica Developer’s Network web site). These metadata integration components
use either the MX SDK or MX Views, an earlier and simpler metadata integration technology
based on database views in the Informatica repository. If your BI tool vendor does not have
an integration offering, then you can use the MX SDK to build your own. The MX SDK comes
complete with sample code to show you how.
Of course, delivering metadata to end-users only has value if the metadata is complete and is
current. Metadata Exchange for Data Models and PowerConnect Metadata Exchange provide
you with tools for capturing the metadata from data models, database catalogs and ERP
systems.
Q: What is PowerPlug?
A: PowerPlug is an earlier name for Metadata Exchange for Data Models (formerly PowerPlug
for Data Models) and PowerConnect Metadata Exchange (formerly PowerPlug for ERP). Some
third-party vendors have also used the term PowerPlug for their own metadata integration
offerings.
Q: What is PowerConnect Metadata Exchange?
A: PowerConnect Metadata Exchange is a proactive metadata management tool for use with
PowerConnect for SAP and PowerConnect for Peoplesoft. Data derived from ERP systems
often underpins a data warehouse implementation but ERP systems are large and complex,
containing thousands of tables and data structures and a change in any one can potentially
cause the data warehouse load to fail. PowerConnect Metadata Exchange is used for
preventative maintenance, to ensure that your Informatica ERP sources are always consistent
with the ERP system that they represent. It quickly and selectively identifies metadata
differences between the ERP system and the Informatica repository, determines the impact of
the discrepancy and can optionally update the Informatica repository to resolve the problem.
Q: Can I use PowerConnect Metadata Exchange with my PowerConnect for Siebel?
A: YES
Q: What is Metadata Exchange for Data Models?
A: Simply put, Metadata Exchange for Data Models is a tool for capturing the metadata that
describes relational databases, then replicating that metadata in the Informatica repository as
either sources or targets. Metadata Exchange for Data Models can capture metadata from the
market-leading database design tools as well as from database catalogs, Informatica
repositories and from XML files that conform to the relational package of the OMG Common
Warehouse Metamodel standard. An interactive and intuitive GUI provides fine-grained control
of metadata replication as well as model validation, reporting and impact analysis.
Metadata Exchange for Data Models saves much time and effort in the development stages of
an Informatica implementation. You design and document your database schema using a data
modeling tool (such as CA ERwin and ModelMart, Embarcadero ER/Studio, Sybase
PowerDesigner or Oracle Designer), then use Metadata Exchange for Data Models to replicate
your data model in the Informatica repository, creating either source or target definitions. As
you refine your data models, database schemas and mappings, Metadata Exchange for Data
Models helps you to detect and resolve metadata discrepancies that may cause your mappings
to fail. Used in conjunction with Informatica business names, descriptions and metadata
extensions, you can document your design with rich metadata, including business and
technical definitions, metadata and data lineage, volumetric data and extended metadata from
3
your data models. For ongoing preventative maintenance and troubleshooting, built-in
validation, change and dependency reports can help to detect and determine the impact of
database schema changes.
Q: I already use a Metadata Exchange tool. Should I upgrade?
A: New versions of Metadata Exchange tools are usually released simultaneously with new
versions of PowerCenter. If you plan to upgrade PowerCenter then you should also plan to
upgrade your Metadata Exchange tools to ensure connectivity and compatibility. This is
particularly important with major releases.
If you are not planning to upgrade PowerCenter there may still be good reasons for upgrading
Metadata Exchange for Data Models to a newer version. For instance, Metadata Exchange for
Data Models 7.1 supports PowerCenter 7.1, is backwards compatible with PowerCenter 6.x and
offers new interfaces for Embarcadero ER/Studio and CA ERwin ModelMart, finer grained
control of comparison reports and a number of bug fixes and minor enhancements. If you use
PowerCenter 6.x then the upgrade is recommended.
Q: What data modeling tools are supported by Metadata Exchange for Data Models?
A: Metadata Exchange for Data Models 7.0 supports CA ERwin 4.1, 4.0 and 3.5.2; CA Model
Manager (ModelMart) 4.1 SP3; Embarcadero ER/Studio 6.0 and 5.5.x; Sybase PowerDesigner
9.5, 9.0, 8.0, 7.5 and 6.1; Oracle Designer 9i, 6i and 6.0 and Oracle Designer 2000 2.1.x.
Support for Embarcadero ER/Studio and CA Model Manager (ModelMart) is new in this release.
Q: I don’t use a data-modeling tool. Do I need Metadata Exchange for Data Models?
A: Apart from data models, Metadata Exchange for Data Models can also extract metadata
from database catalogs, Informatica repositories and CWM files.
The database catalog interface provides a powerful tool for detecting discrepancies between
Informatica sources and targets and the corresponding database schema. If you detect
changes, you can update the Informatica repository from the database schema, guaranteeing
consistency.
If your database platform supports schema comments (e.g. DB2, Oracle), then you can also
capture these in the Informatica repository.
Using the Informatica repository interface, you can connect to two different Informatica
repositories and compare sources and targets, transfer business metadata data and migrate
sources and targets between repositories.
The XMI: Common Warehouse Metamodel interface provides a standards-based mechanism
for metadata exchange, using the relational package of the OMG Common Warehouse
Metamodel standard.
Q: Informatica Designer can import database schemas. Why do I need the database
catalog interface in Metadata Exchange for Data Models?
A: Informatica Designer can use ODBC to create source and target definitions from a database
catalog. However, if you already have the source or target in the Informatica repository,
Designer has to delete the existing object, before creating a new one – this invalidates your
mappings. Metadata Exchange for Data Models can compare the database catalog with the
sources and targets, report on differences and selectively create or update the corresponding
Informatica repository objects.
Metadata Exchange for Data Models also gives you far more
control over the business metadata. It can capture database comments (if the database
supports them), can generate business names from table and column names, and can
selectively update just the physical metadata while leaving existing business metadata and
mappings intact. If you define suitable extended metadata properties, you can also capture
auditing information including the source database catalog and an update timestamp.
4
Q: Why would I use Metadata Exchange for Data Models to move metadata between
Informatica repository folders when I can do this with Informatica client tools?
A: Informatica Designer allows object definitions to be exported as XML files and allows the
XML files to be imported. Repository Manager provides tools for copying folders between
repositories. Metadata Exchange for Data Models complements these tools by allowing
conversion of sources to targets and targets to sources (useful for staging tables and
Operational Data Stores) and also allows migration between versions. Metadata Exchange for
Data Models also allows selective migration of business names, descriptions and metadata
extensions and mapping between them – this provides a very useful tool for managing
business metadata and for moving business metadata between development, test and
production environments.
Q: Can I update my data models from my Informatica repository?
A: No. Data model interfaces are currently read-only and metadata exchange occurs from the
data model into the Informatica repository. The reasons for this are primarily technical: until
recently, data modeling tools did not provide an API that allowed the data model to be
updated.
Another significant reason is the need to define a metadata system of record: generally this is
the data model.
However, you can select two different data models at one time; all of the reporting features of
Metadata Exchange for Data Models are available. This means that you can compare two
different data models, even if the data models were created with different modeling tools. You
can also select a data model and a database catalog and compare the data model with the
database schema that it represents.
Q: Do any interfaces support bi-directional metadata exchange?
A: Yes. The Informatica repository interface is bi-directional and it can be used as both a
metadata source and a target. The Common Warehouse Metamodel interface is also
bi-directional and can be used for metadata exchange with any tool that supports the standard
and the relational package.
Q: What Informatica repository objects do the Metadata Exchange tools support?
A: Metadata Exchange for Data Models can selectively update both Informatica source and
target definitions. It can update all of the object’s properties, including table, column and
index names, physical properties (e.g. column datatypes), business names, descriptions and
metadata extensions.
PowerConnect Metadata Exchange works specifically with Informatica source definitions for
ERP sources (SAP and Peoplesoft). It can update the physical metadata, ERP specific
properties, business names and descriptions.
Q: What is the CWM standard, and how do the Metadata Exchange tools support it?
A: The Common Warehouse Metamodel is a vendor-independent metadata standard defined
by the Object Management Group OMG). The CWM standard defines the key metadata
required for data integration and defines a file format and mechanism for exchanging it. This
standard supercedes the earlier Metadata Coalition MDIS standard and is supported by many
key players in the industry. Informatica and Yaletown Technology Group are both OMG
members.
Informatica Metadata Exchange for Data Models provides an interface that can read and write
XML files that conform to the CWM relational package. This means that metadata can be
exchanged with any other product that implements the standard and contains relational
database metadata.
5
Q: Can I run the Metadata Exchange products in batch mode?
A: Because of the large number of objects in an ERP system, PowerConnect Metadata
Exchange was designed specifically to support batch operation. A command file controls batch
operation. The command file is configured using an interactive tool and specifies the ERP and
repository connection parameters, the ERP sources in the Informatica repository that will be
included in reports and updates, and the operation that will be performed. The name of the
command file is specified as a parameter when the batch operation is started.
Metadata Exchange for Data Models cannot be run in batch mode but customers have asked
for this feature, so it may be offered in the future.
Q: How is Metadata Exchange for Data Models licensed?
A: Separate licenses are available for ERwin, ModelMart, ER/Studio, PowerDesigner and
Oracle Designer. These licenses cover all versions of the modeling tool, even where different
integration technologies are used. Special licensing packages are available for the CWM,
database catalog and Informatica repository interfaces. Enterprise licensing for single or
multiple sources is also available.
Q: How is Metadata Exchange for Data Models packaged?
A: Metadata Exchange products can be downloaded from the Informatica Developer’s Network
web site. The Metadata Exchange for Data Models download is a trial version that permits a
single use per interface. A license key must be purchased to unlock the full product
functionality.
Q: How much does Metadata Exchange cost?
A: The Metadata Exchange SDK is available to all Informatica customers from the Informatica
Developer’s Network. For the latest pricing and more information on Metadata Exchange
products, please contact your Informatica sales representative.
Q: How are the Metadata Exchange tools supported?
A: Metadata Exchange tools are Informatica products. To obtain support and product updates,
you need to purchase support from Informatica. Additional resources are available on the
Informatica Developer Network web site.
Q: What operating systems does Metadata Exchange run on?
A: Metadata Exchange products all use the MX SDK. The MX SDK is built using Microsoft's
Component Object Model (COM) standard and requires 32bit Windows operating systems
including Windows 98, Windows NT 4, Windows 2000 and Windows XP.
Q: What repository database platforms does Metadata Exchange support?
A: Metadata Exchange works with all of the repository database platforms that are supported
by the Informatica Data Integration platform. Since version 6, the Metadata Exchange SDK
uses TCP/IP to interact with the Informatica Repository Server, providing complete database
platform independence, enhanced security and better performance.
Q: Does Metadata Exchange for Data Models support all the database platforms that
my data modeling tools do?
A: ERwin ER/Studio and PowerDesigner are designed to support almost all relational database
platforms. Metadata Exchange for Data Models supports almost all of the database platforms
that these products support, including all of the popular ones. Informatica PowerCenter has
native support for Oracle, Sybase ASE, Informix, DB2, Microsoft SQL Server and Teradata.
Metadata Exchange for Data Models maps data models defined for these database platforms to
the corresponding Informatica native database type and all others to ODBC (e.g. Microsoft
Access 2000, Red Brick, Centura SQLBase, Sybase SQL Anywhere).
6
Oracle Designer provides an Oracle centric modeling solution. Oracle Designer Server Model
Diagrams do not explicitly specify a database platform, so Metadata Exchange for Data Models
allows the user to choose the database platform and maps Oracle Designer datatypes to the
corresponding database platform specific datatype.
Q: Does Metadata Exchange support XML?
A: Metadata Exchange products use XML for metadata exchange, reporting and control.
Metadata Exchange for Data Models can read and write XML files that conform to the relational
package of the OMG Common Warehouse Metamodel standard, allowing metadata exchange
with other tools that support the standard.
The MX SDK can be used to export and import XML files that conform to the Informatica
repository DTD. This feature utilizes functionality that is built into the Informatica Repository
Manager, Designer and PowerConnect, producing files that can be exchanged between
Informatica tools. PowerConnect Metadata Exchange uses the export functionality to generate
XML files from both the ERP system and the Informatica repository and uses the import
functionality to update the repository.
Metadata Exchange for Data Models and PowerConnect Metadata Exchange use proprietary
XML file formats to store report data and XSL style sheets to transform the data for
presentation. PowerConnect Metadata Exchange command files are also XML files.
ETL Tool vs EAI Tool
An ETL tool takes data from one or more data models (typically represented by an EntityRelationship diagram) and puts it into another data model (represented by another E-R
diagram). As the name implies, these models represent entities -- invoices, people, budgets,
even timelines – and their relationships. The order in which things happened (usually) isn't all
that relevant to a data model. You could say that an ETL tool coordinates multiple entities and
their relationships as data objects.
But an EAI tool coordinates multiple entities and their relationships within a given process. We
deal with process flows and the transactions (not so much individual entities) within that
process. The order in which things happen are critical, and the relationship of entities to each
other is less important than the relationship of transactions to each other.
Clearly, these two can be related. You can occasionally do ETL with an EAI tool, and vice
versa.
7
Tips for interview:
Interviews are a lot of work and require serious preparation. Review your
recent performance and have examples of how you
1) solved a complex issue,
2) displayed leadership,
3) exhibited team spirit. Focus on accomplishments.
Review in detail the requirements of this new post.
Wear a nice suit and be clean-shaved.
Anticipate possible questions and have some well prepared responses. Be
ready to ASK GOOD QUESTIONS.
2. Adapted from response by Joe on Wednesday, December 15, 2004
Being nervous is natural, especially for an important experience you are about to go
through. Some suggestions:
- Review the company, the division and the people you are going to work for. Learn as much
about them, their products, their vision, their mission, etc... Study online resources for this
information, but also get on the phone, even seek meetings (informal ones) with others in the
company, or with those who know the company. This is considered a normal/natural part of
your job seeking homework, by the way. Learn all you can, as it’s in your best interest.
- Have one, or more, people who you know/trust to give you a series of mock
interviews. What worked for me in this area (your mileage may vary!) was for my mock
interviewer to set up a series of 3-5 interviews, each was to be a new/unique session and to
put me through a variety of typical scenario's. I was fortunate that this was a person who does
this for a living. I was blunt and candid in what I wanted, and expected to be shown what a
nice/good/effective interview SHOULD be like, but to also pointedly put me on the spot with
how things can go wrong. Each session was treated like a full and FORMAL job interview, to
include suit, demeanor and complete interview set of questions, answers and
discussion. Afterward there was a blunt and candid review of what the objectives for that
session were (from HIS perspective), what areas I did well on, where I did poorly/badly,
identification of areas of opportunity where I missed out on something good (or bad) to
capitalize on, and objective suggestions for improving my body language, demeanor,
language, and attitude. In my mind, I wanted these mock interviews to take their best shots
at ripping me to shreds, and see where my strong/weak points were. The reviews afterward
were essential to improving my understanding of MYSELF and what I MUST improve in order
to get through the interview. For me, this proved to be a winning move.
- Go into the interview eager and ready to experience it. Relish and enjoy every moment of it.
You will get to do it so infrequently, that this is a golden opportunity to experience to the
fullest. You may think I'm kidding ---I'm not. By adjusting yourself so that this IS your
mindset and approach, you'll find it not only enjoyable, but very rewarding as well.
- I'm hoping someone more knowledgeable than me can address the interview questions
'issue'. I know there are lots of resources on the internet to research the plethora of interview
questions and types, etc... My suggestion, at this point in your life/career, to not worry so
much about the questions, as to what you can give and offer this company. On the other
hand, you certainly should have your own list of questions, written down is fine, of what you
want to know about them. Especially about your work environment, expectations of you and
your time, etc... Start off general ("what can you tell me about the company") and work to
being more specific ("what can you tell me about the division", what can you tell me about the
position you are hiring me for", etc....").
8
- Try this approach on being calm---think about, and continuously remind yourself in
productive, enriching and positive ways that you will calmly and rationally be successful in this
interview. Mentally focus on what you WANT, vice what you don't want. It’s fine to honestly
self-evaluate how you are today. What is really important is HOW will you improve? What can
you do better, and what are you doing about it now? Another approach is that being nervous
is your minds way of telling you to be careful. You are in control of yourself. You decide what
is important or worrisome. So, tell your mind what to think and how to act. Such an
improvement can occur over time when you are persistent. Think about it. Side note: I've
found in life, people who focus on what they don't want, or like, as the case may be, don't see
how negative that is. They really believe that by telling themselves NOT to do something that
somehow, magically, the RIGHT thing they are supposed to be doing will magically occur. It
doesn't work that way. I've found when you positively and actively WANT something to occur,
then make that accomplishment the focus of your attention --- it happens. I believe that
occurs because you've DONE something, as opposed to the alternative of attempting to NOT
do something. I believe the former is a positive builder in our lives. Be optimistic.
- You gave the impression that you'd have some stiff competition for this job. Competition is a
good thing. Go in with your best foot forward. Be honest, show them that you are more than
interested in doing your best every day. Convey, throughout the interview-using every
question as an opportunity, to explain to them HOW dedicated you are, HOW energetic your
day-to-day
performance is now (and will continue to be for them), that you are the best candidate even
with your 'limited' experience (because you can work harder, smarter, faster, with stellar
results), that you are a quick learner, that you are able to efficiently and effectively apply new
rules to existing paradigm's, etc.... In accomplishing this, I don't mean for you to embelish
who and what you are, simply have this (the above) as your mindset, and as a way to
EXPLAIN/justify and demonstrate that YOU are confident you are the best candidate. Note: In
this context, I mean no offense with the reference to your being 'limited', simply saying that
your skills are what they are, and that it is normal to recognize others (your competition?)
have more time, and possibly even more talent/skill than you do. In fact, consider the amount
of your skills, and that you are positively approaching this job opportunity as an 'asset'. You
are bringing knowledge and experience to the table. You'll accept direction on what they want,
and how they want it, but will also tap into your own talents, skills, abilities and creativity to
do the job even better. Review your resume, the one THEY have, the night before. Study it
closely, as they will ask you questions based on what they see, and don't see. I've seen a
good interview go VERY bad, simply because the interviewee wasn't cognizant of the content
of their own resume. Focus on your talents and skills. Don't BS the interviewer with smoke
and mirrors. More than likely they'll catch on rather quickly that you are smart, patient and
honest (desired qualities!!!) or that you aren't (bad).
- Be honest regarding what you can do, and only volunteer what you are bad at, or cannot do
when questioned about something specific that you can't do. Its reasonable to know your
limitations, and that you can candidly explain the breadth of your abilities (and limits). If you
find they focus on 'stuff' you don't know, its ok. Expect such questions and take them in
stride. Follow up with your speed/willingness to learn AND apply it correctly.
- Finally, as quickly as possible after the interview, sit down (in your car even) and take notes
on what occurred. What could you have done better? What things did you mention or explain,
that your resume SHOULD have addressed? What did you do right, and can capitalize on in
the future? What questions were asked, and how you answered them. Its normal that you
won't remember everything, yet by documenting the gist of it when it is fresh in your mind,
will be productive and helpful when you go through, and prepare for, the next interview. You
think you were going to stay in this new job for forever, did you?
Do NOT take it personally, if you don't get this job. Instead, reflect on this interview, work on
areas you can improve on, and move on to seeking out the next one. More than likely, the
company hired the best candidate, and that honestly will not always be you. That is ok and
should be acceptable in your mind. Such things happen, and companies make job-hiring
9
decisions in light of what is best for the company. It greatly helps to realize that these are
business decisions and not an adverse indicator of you (which it isn't!).
3. Adapted from response by Shareen on Wednesday, December 15, 2004
As an advice on interviews, the best way is to be honest in your answers. If you have never
worked on the task that is asked on the interview, the best answer is to say I can learn
anything quickly even though I have never worked with what they ask I have a proven track
record on learning new tasks quickly. It is important to bring the positive first and not the
negative. Example, I never have worked with the software but I can learn quickly. See, in the
example, the interviewer will only hear the negative and not the ending part of the sentence.
Another word of advice, be yourself. It is not advisable to rehearse answers on a face to face
interview. The reason is that the interviewer will pick up on it and try to trip you up or not
believe your answers.
You should also be relaxed. Remember the person who is interviewing you must want to talk
to you and like what you have to offer or they would have never agreed to an interview. You
should also remember that the person interviewing you has been in your position before and
will understand some anxiety.
Consider what specifically your need is.
Typically an ETL tool
1. reads data
2. may change data
3. writes data
In a majority of cases (50-80%) it just reads and writes to a new location.
In
1.
2.
3.
4.
5.
the rest of the cases it does a 'simple' transformation text to numeric
text to date
format - leading '0', dashes etc.
simple split - one field to two or more fields
simple merge - two or more fields to one (name concatenation)
In
1.
2.
3.
4.
a few cases it does more
look up to another table and replace
complex algorithm
merge two or more input fields into one with some algorithm
split one field into many with some algorithm
Breaking all this down you can create a set of procedures that will
1. read a database for input information
2. write that information to file
3. read a database for transformation code
4. append that code to file
5. read a database for write/output information
6. append that information to file
7. execute file
You've created your ETL. This process will probably handle 80-95% of all your needs. The
other 5-20% will need specific coding, but may (if you pay attention) fall into several sets of
code that, if you analyze the code you may be able to insert that code into your 'ETL' database
and then call it as needed.
b>More information:
10
More information on Data Warehouse solution selection is available in the DW-SELECT
discussion group. Enter a topic or subject and click the search button for detailed results.
General Questions on data ware housing:
Here a few:
1.
2.
3.
4.
5.
Tell me about cubes
Full process or incremental
Are you good with data cleansing?
How do you handle changing dimensions?
What is a star-schema?
2. Adapted from response by Mike on Thursday, July 22, 2004
A few high level questions might include:
Talk about the Kimball vs. Inmon approaches.
Talk about the concepts of ODS and information factory.
Talk about challenges of real-time load processing vs. batch.
For Informatica:
Let them know which version you are familiar with as well as what role. Informatica 7.x has
divided the developer and administrator roles.
You will most likely be asked specific questions for building a mapping and workflow. Know
what the difference is between sttic and reusable objects for both. Be prepared to
demonstrate how to create a connection, source definition (flat file and relational), use
expression transformation, lookups (connected and disconnected), aggregators, normalizers,
update strategies, how to modify source and target sql overrides, etc.
For Erwin:
Know the difference between Logical and Physical models.
Know how to use the Reverse Engineer and Comparison features.
The dimension model feature is pretty weak, but you might want to know how Erwin treats
dimensional modeling.
Other topics:
Anything you know about RDBMS is worth discussing.
In Oracle, you can talk about referential integrity as it applies to DW. Views and Materialized
Views, Partitioning, Bitmap Indexing (when to use), and any other specifics as related to DW
(for 10g there is the new Bitmap join Index).
Always, always offer details of your knowledge, and ask questions to get the customer's
perspectives (you do not want to push Kimball concepts if the customer is hard-set on Inmon).
Anything you can bring to the table regarding the customers business systems (i.e. SAP,
Peoplesoft, etc.) will help separate you from the pack. Also anything you know about business
processes such as:, Order Fulfillment, Inventory Analysis, Finance, etc. will also separate you.
3. Adapted from response by Shaquille on Wednesday, August 11, 2004
Here are the few questions that might be posed:
Data Warehousing questions:
1) What is source qualifier?
11
2) Difference between DSS & OLTP?
3) Explain grouped cross tab?
4) Hierarchy of DWH?
5) How many repositories can we create in Informatica?
6) What is surrogate key?
7) What is difference between Mapplet and reusable transformation?
8) What is aggregate awareness?
9) Explain reference cursor?
10) What are parallel querys and query hints?
11) DWH architecture?
12) What are cursors?
13) Advantages of de normalized data?
14) What is operational data source (ODS)?
15) What is meta data and system catalog?
16) What is factless fact schema?
17) What is confirmed dimension?
18) What is the capacity of power cube?
19) Difference between PowerPlay transformer and power play reports?
20) What is IQD file?
21) What is Cognos script editor?
22) What is difference macros and prompts?
23) What is power play plug in?
24) Which kind of index is preferred in DWH?
25) What is hash partition?
26) What is DTM session?
27) How can you define a transformation? What are different types of transformations in
Informatica?
28) What is mapplet?
29) What is query panel?
30) What is a look up function? What is default transformation for the look up function?
31) What is difference between a connected look up and unconnected look up?
32) What is staging area?
33) What is data merging, data cleansing and sampling?
34) What is up date strategy and what are th options for update strategy?
35) OLAP architecture?
36) What is subject area?
37) Why do we use DSS database for OLAP tools?
Business Objects FAQ:
38) What is a universe?
39) Analysis in business objects?
40) Who launches the supervisor product in BO for first time?
41) How can you check the universe?
42) What are universe parameters?
43) Types of universes in business objects?
44) What is security domain in BO?
45) Where will you find the address of repository in BO?
46) What is broad cast agent?
47) In BO 4.1 version what is the alternative name for broadcast agent?
48) What services the broadcast agent offers on the server side?
49) How can you access your repository with different user profiles?
50) How many built-in objects are created in BO repository?
51) What are alertors in BO?
52) What are different types of saving options in web intelligence?
53) What is batch processing in BO?
54) How can you first report in BO by using broadcast agent?
55) Can we take report on Excel in BO?
12
1)what is the use of surogate keys?
surrogate key in a data warehouse is more than just a substitute for a natural key Surrogate
keys are the artifical keys that are used in SCD
2)where to use dd_insert and insert as update?
dd_insert will insert data, where as insert as update will be used in SCD concept.
3)what is the use of updatestratage transformation?
Update stratery transformation will look for all the updates,inserts and deletes.
4)what is associate port in dynamic lookup?
hope i told u this
5)what is sql design?
6)why sorter tra is active?
Sorter transformation is generally used to sort the data. it Have an option called distinct in
properties sheet this option will remove the duplicates. so once if we pass 10 records having 5
duplicates it will give only 5 records out which tells this as a active
7)how to improve query optimization?
rule based
cost based
it depends on requirment
8)what is meant by globel enterprice repository?
9)how to compare ports in expresion (other than warehouse key)
10)what types of document your are getting from client?
11)what is materlized view?
A materialized view is a database object that contains the results of a query. They are local
copies of data located remotely, or are used to create summary tables based on aggregations
of a table's data. Materialized views, which store data based on remote tables are also, know
as snapshots.
12)what is referential cursor?
13)what advantage of denormalization? and what is denormalization?
In DWH we store historical data if the tables are normalized it will be difficult to
14)in session level sql overwrite is there or not?
15)what is data modular?
16)suppose u have 12 tran in one mapping but the data is not flowing the 3rd trans
how to know what is the problem?
17)how many mapings are done in your project?
13
depends on duration of project
18)what is workflow variable?
we can specify when u create a work flow these variables are generally used while executing
the mapping
19)currently where u r working in your project?
20)how to improve performance in maping level?
hope u know this
21)how to send post e-mail to client if session is not completed?
22)what is parrel partioning?
hope u know
23)suppose you have 6 sessions in your mapping how to identify which typ of batch
is running in that mapping?
we can see it in workflow monitor
24)i have implemented one mapping with one source and one target later my client
given two more sources what is the initial step you have to do?
nothing much load those table also in mapping if there is any requirment like first one target
shd get data populated before the other we can sprecify that in target load plan
14