ETL vs DATA Integration

[MG18]

Hi All,

what is the difference between ETL and data integration and SAS Data integration? or both (ETL DI) are the same ?

Regards,

[Kurt_Bremser]

---

@MG18 wrote:

Hi All,

what is the difference between ETL and data integration and SAS Data integration? or both (ETL DI) are the same ?

Regards,

---

ETL is the generic name for the process of Extract - Transform - Load that describes the path for data from the external environment into a data warehouse.

Data Integration is the name for SAS tools designed to support the user in the ETL process, mainly the Data Integration Server and Data Integration Studio.

[Patrick]

@Kurt_Bremser

Data Integration is as much a common industry term as ETL. Nothing that SAS invented.

I personally prefer to use the term Data Integration over ETL as it "feels" a bit broader to me and avoids me getting into ETL vs. ELT discussions.

[MG18]

Hi All,

My question is is there any difference between the ETL and data integration ? or both are the same ?

[TomKari]

And of course you have to differentiate between "data integration" and "ETL" being used as general terms, not specific to a vendors products, as opposed to a vendor who uses either term in one of their products, such as "SAS Data Integration Studio".

In the former case, I consider them equivalent. In the latter, it of course means whatever the vendor wants it to mean!

As general terms, in my opinion, they both refer to the process of preparing data to be stored in a "clean, production" environment, usually by editing and transforming it from its raw and rather messy state. ETL of course conceptually implies that the source and the sink are databases, but as a general term it's used in the wider sense.

My 2 cents worth,

   Tom

[LinusH]

For me, ETL as a term just tells you that you copy, transform and load data somewhere.

Integration on the other hand, implies that we have a greater cause, and that is actually integrating data (and that isn't just putting different data into the same platform). It can be have a common naming convention, attribute definitions, shared data model, and at the far end integration of business keys and domain implementing global domain values.

[GertNissen]

It's all Data Management 😉

Data Integration is a broad term describing how to "work" with data, i.e. building a Data Warehouse - for that you can use different load techniques - such as ETL, ELT, DataVault, StarSchemes, MapReduce, UPM (Unified Programming Model) etc. etc.

Sometimes Data Quality is an "add-on" to the ETL, to emphasize the extra importance (others might argue its part of the T in ETL). Perhaps you could skip the T and pour all data into a Data Lake? How would ETL be possible in Real Time environments?

Some would might also add Data Migration and Data Consolidation as additional processes to Data Integration or Data Federation/virtualization. Master Data Management could also be added into the same context.

There is an short SAS article here that has an good explanation https://www.sas.com/da_dk/insights/data-management/what-is-etl.html

Gartner uses this definition
The discipline of data integration comprises the practices, architectural techniques and tools for achieving the consistent access and delivery of data across the spectrum of data subject areas and data structure types in the enterprise to meet the data consumption requirements of all applications and business processes.
Source: https://www.gartner.com/it-glossary/data-integration-tools/

SAS has many Data Management tools - https://www.sas.com/da_dk/solutions/data-management.html?gclid=CjwKCAjw3qDeBRBkEiwAsqeO7jwzz1tZn7Ah7...

In SAS you would typically use the tool SAS Data Integration Studio (+ Data Management Studio) to build your ETL jobs/flows to build your Data Warehouse.