Presentation Determination

8/4/2019 Presentation Determination

1/39

Determination

Presented By :

Rupal Bhatt

Mikky Mahant

Jateen Makawana

Sanjay Dubey

1


2/39

2


3/39

Ports

All input / outputSpecify a Filter Condition

UsageFilters rows from inputflow

Filter Transformation

Drops rows conditionally


4/39

Can sort data from relational tables or flat files

Sort takes place on the Informatica Server machine

Multiple sort keys are supported

The Sorter transformation is often more efficient than a sortperformed on a database with an ORDER BY clause

Sorter Transformation


5/39

contd

PortsAll input / outputDefine one or more sort

keysDefine sort order for eachkey

Example of UsageSort data beforeAggregator to improveperformance

Sorts data from any source, at any point in a data flowSort Keys

Sort Order


6/39

Cache size can beadjusted. Default is8MbServer uses twice thecache listed

Ensure sufficientmemory is available onthe Informatica Server(else the Session Taskwill fail)

Properties

contd


7/39

PortsMixedVariables allowedGroup By allowedCreate expressions in outputports

UsageStandard aggregations

Aggregator Transformation

Performs aggregate calculations


8/39

AVGCOUNTFIRSTLASTMAXMEDIAN

MINPERCENTILESTDDEVSUMVARIANCE

Functions

Return summary values for non-null data in selected portsUse only in AggregatortransformationsUse in output ports only

Calculate a single value (and row)for all records in a groupOnly one aggregate function can benested within an aggregate functionConditional statements can be used

with these functions

contd


9/39

Instructs theAggregator to expectthe data to be sorted

Set Aggregator cachesizes for InformaticaServer machine

Properties

Sorted Input Property

contd


10/39

Sorted Data

Aggregator can handle sorted or unsorted dataSorted data can be aggregated more efficiently, decreasing totalprocessing time

The Server will cache data from each group and release thecached data

upon reaching the first record of the next group

Data mustbe sorted according to the order of the Aggregators

Group By ports

Performance gain will depend upon variable factors

contd


11/39

Data Flow Rules - Terminology

Passive Transformation- Operates on one row of data at a time AND- Cannot change the number of rows on the data flow- Example: Expression transformation

Active Transformation- Can operate on groups of data rows AND/OR- Can change the number of rows on the data flow- Examples: Aggregator, Filter, Source Qualifier


12/39

Active Transformation

Ports

All input or input / outputM denotes port comesfrom master source

ExamplesJoin two flat filesJoin two tables from

different databasesJoin a flat file with arelational table

Joiner Transformation

Performs heterogeneous joins on different data flows


13/39

Joiner Conditions

Multiple join conditionsare supported

contd


14/39

Joiner Properties

Join types:Normal (inner)Master outer

Detail outerFull outer

Set Joiner Caches

Joiner can accept sorted data (configure the join condition to use the sort origin ports)

contd


15/39

For each mapping row, one or more port values are looked up in adatabase table or flat file

If a match is found, one or more table values are returned to themapping. If no match is found, NULL is returned

Lookup Transformation

Lookup valuesLookup transformation

Return values


16/39

Ports

Mixed

L denotes Lookup port

R denotes port used as areturn value (forunconnected Lookups)

Specify the Lookup Condition

Usage

Get related values Verify if a record exists or if

data has changed

Looks up values in a database table or flat file andprovides data to other components in a mapping

contd


17/39

Lookup Conditions

Multiple conditionsare supported

contd


18/39

Lookup table name

Lookup Condition

Native database connectionobject name

Source type(database or flat file)

Lookup Properties

contd


19/39

Policy on multiple match:

Use first value Use last value

Report Error

Lookup Properties

contd


20/39

Caching can significantly impact performance

Cached

Looked table data is cached locally on the server

Mapping rows are looked up against the cache

Only one SQL SELECT is needed

Uncached

Each Mapping row needs one SQL SELECT

Rule Of Thumb: Cache if the number (and size) of records in theLookup table is small relative to the number of mapping rowsrequiring the lookup

Lookup Caching

contd


21/39

By default, Lookup caches are not persistent; when the session

completes, the cache is erased

Cache can be made persistent with the Lookup properties

When the Session completes, the persistent cache is stored on theserver hard disk

The next time the Session runs, cached data is loaded fully orpartially into the RAM and reused

A named persistent cache may be shared by different sessions

Can improve performance, but stale data may pose a problem

Persistent Caching

contd


22/39


23/39

Make cache persistent

Set Lookup cache sizes

Set prefix for persistentcache file name

Reload persistent cache

contd


24/39

To maintain referential integrity, primary keys must be loadedbefore their corresponding foreign keys: in order tgt1, tgt2, tgt3

Constraint-based Loading

PK1

PK2

PK2

PK3

PK3

contd


25/39

Setting Constraint-based Loading

contd


26/39

Active transformation

Can operate on groups of data rows and/or can change the number ofrows on the data flow

Examples: Source Qualifier, Aggregator, Joiner, Sorter, Filter

Active Source Active transformation that generates rows Cannot match an output row with a distinct input row Examples: Source Qualifier, Aggregator, Joiner, Sorter NOTE: The Filter is NOT an active source

Active Group Group of targets in a mapping being fed by the same active source

Constraint-based Loading - Terminology

contd


27/39

Example 1

With only one Active source,rows for targets 1, 2, and 3will be loaded properlyand maintain referentialintegrity

Example 2:

With two Active sources, it isnot possible to controlwhether rows for tgt3 willbe loaded before or afterthose for tgt2

Constraint-based Loading - Restrictions

Cannot have two active groups

PK1

PK2

PK2

PK3

PK3

PK1

PK2

PK2

PK3

PK3

contd


28/39

Ports

All input / output

Specify the UpdateStrategy Expression IIF or DECODElogic determineshow to handle therecord

Example Updating Slowly

ChangingDimensions

Update Strategy Transformation

Used to specify how each individual row will be used toupdate target tables (insert, update, delete, reject)


29/39

IIF ( score > 69, DD_INSERT, DD_DELETE ) Expression is evaluated for each row

Rows are tagged according to the logic of the expression

Appropriate SQL (DML) is submitted to the target database:insert, delete or update

DD_REJECT means the row will not have SQL written for it. Thetarget will not see that row

Rejected rows may be forwarded through Mapping

Update Strategy Expressions

contd


30/39

Router Transformation

Sequence Generator Transformation Normalizer Transformation

Rank Transformation

Union Transformation

Other Transformations


31/39

Mapplets are created in the mapplet designer

Mapplets are used when you want to use a standardizedset of transformation logic in several mappings When you use a mapplet in a mapping only the input and

output ports from the transformations are viewed

Mapplets

Mapplet


32/39

Mapplets help simplify mappings in the following ways: Include source definitions. You can include multiple source

definitions and qualifiers to provide data for a mapping.

Accept data from sources in a mapping. If you want the mapplet toaccept data from a mapping you can use an input transformation to

receive source data. Include multiple transformations. A mapping can contain as many

transformations as you need.

Pass data to multiple transformations. You can create a mapplet tofeed data to multiple transformations in the mapping.

Contain unused ports. You do not have to use all input and outputports in a mapplet.

Reusable. A complex mapplet can be used in multiple mappings

contd


33/39

Useful for repetitive tasks / logic

Represents a set of transformations

Mapplets are reusable

Use an instance of a Mapplet in a Mapping

Changes to a Mapplet are inherited by all instances

Server expands the Mapplet at runtime

Mapplet Advantages

contd


34/39

Do not use the following in a mapplet:

XML source definitions

Target definitions Other mapplets

Unsupported Transformations

contd


35/39

Internal Sources

One or more Source definitions / Source Qualifiers within theMapplet

External Sources Mapplet contains a Mapplet Input transformation

Receives data from the Mapping it is used in

Mixed Sources

Mapplet contains one or more of either a Mapplet Inputtransformation AND one or more Source Qualifiers

Receives data from the Mapping it is used in, AND from theMapplet

Mapplet Source Options

contd


36/39


37/39

Define once, use many times Can be a copy or a shortcut

Edit Ports only in Transformation Developer

Can edit Properties in the mapping

Instances dynamically inherit changes

Caution: change reusable transformations can invalidatemappings

Note: Source Qualifier Transformations cannot be made reusable

Reusable Transformations


38/39

This copy action must be done within the same folder

Hold down the Ctrl key and drag a Reusable transformation fromthe Navigator window into a mapping (Mapping Designer tool)

A message appears in the status bar:

Make a non-reusable copy of this transformation and add it to thismapping

Drop the transformation into the mapping

Save the changes to the Repository

Copying Reusable Transformations

contd


39/39

Thank You

Presentation Determination

Documents