|
Speed
ETL loads all your corporate
data, cleanses the data with the rule specified, andthen loads
the data into Speedminer repository - Speed OLAP. You can
be assured that your data in Speedminer OLAP is reliable,
complete, accurate, integrated, up-to-date and ready for your
Business Intelligence Analysis.
Speed
ETL with its innovative and powerful design supports calculated
field, customized cleansing rule even the most complicated
one. Speed ETL supports Real Time Data Warehouse via Record
Comparison, Ascending Index, Custom Log, as well as Full Log.
System generated codes
are customizable via VB Script or Object Script.
Speed
ETL is able to capture metadata from most popular DBMS and
thus simplifies the entire ETL process. Speed ETL also supports
creation of Surrogate key and foreign key to ensure integrity
between tables of heterogeneous data sources and can be established
easily.
.jpg)
Connectivity
Speed ETL connects to all popular relational databases,
unstructured data source like MS Excel, flat files, delimited
text files and other common data file formats. Custom connectivity
can be built for legacy application.
On Demand Connection
Speedminer offers the broadest on demand and event trigger
connectivity (more than 250 pre-built adapters) to proprietary
ERP applications (e.g. SAP, PeopleSoft, Oracle, Siebel, JD
Edwards, Manugistics, BAAN etc.), databases and mainframes
via the Ensemble Universal Integration
Platform.
Cleansing
Data quality problems like missing information, out-of-range
values and integrity error can cause inaccurate BI analysis.
Speedminer provides many options for cleansing data to correct
problems with source data before it is loaded into Speedminer
OLAP repository.
Even after ETL
process, Speedminer supports a second level of data cleansing
in Business Terminology Setup Module.
Speedminer's cleansing methods include the following:-
Transformation
and Standardization: Field data can be transformed from
an abbreviated code to a full description or vice versa. Standardization
can also be easily achieved by mapping multiple different
values to a single standardized value.
Ranges Grouping:
Values fall in certain range can be easily grouped and renamed.
Null value replacement:
Empty value fields can be replaced either with a default value
or the return value of a more complex algorithm.
Integrity Error:
Invalid foreign key can be replaced with a default key or
determined by a complex algorithm.
Complex Algorithm:
Speedminer allows users to write customized VB scripting or
Object Scripting code to support most complicated cleansing
rule.
Integration
Data from multiple disparate sources that has no relationship
between them can be linked in Speed ETL.
Scheduler
The Scheduler allows the ETL process
to be carried out at a pre-scheduled time or repeat execution
hourly, daily , weekly etc.
Speed ETL delivers:
Extremely High
Performance
The bottleneck of most ETL appears on the inserting/updating
the data to the staging database. Speedminer bypass SQL to
perform internal fast checking & data manipulation at
lower level which greatly improve the performance of ETL process.
Shortest Implementation
With the elimination of cumbersome cube, Speedminer cuts down
the implementation time of data warehouse project from months
to weeks. Speedminer is the only DW to support OLTP architecture
in OLAP design.
Slowly Changing
Dimension (SCD)
Complex feature like SCD is easily supported in Speedminer
without any hand coding. In addition to Type I - III SCD,
Speedminer supports enhanced Type IV version which gives more
flexibility and power to track changes in data
Multiple Database
Consolidation
Consolidates tables of similar architecture on multiple databases
and automatically inserts SiteID as part of the primary key
to identify source data. Speed ETL also supports auto consolidation
of tables from single/ multiple databases.
Multi-Sites/Multi-Table
Auto consolidation of tables from multiple tables and multiple
databases
E-Mail
Besides status and error of ETL that can be sent via e-mail,
Speed ETL also supports monitoring of data (Business Activity
Monitoring) to e-mail alerts to recipient.
Data Analysis
Speed ETL provides many statistical analyses(means/std.dev/variance/sum/average/skewness/std.error/missing
value/kurtosis/coeff var/uniqueness etc.) of data bsefore
or after loading.
Field Creation
Additional fields with computation or storage can be created
easily.
Functions
Library
Hundreds of statistics/arithmetic/string/date/time etc. functions
are available for cleansing and data manipulation in Function
Library.
|
