Attunity has been a successful player in data space for almost 20 years and well-known for its data replication solution Attunity Replicate, in 2016 brought to the market a data warehouse automation platform Attunity Compose.
Data warehouse automation is a relatively new approach to data warehouse development in a way similar to the DevOps concept in software development. It facilitates the configuration and management of enterprise data warehouse, increases the agility and speed of delivery for analytics solutions to business and reduces project costs. Data warehouse automation encompasses the core processes from design, development, testing, deployment to operation support and change management.
Here is a brief guide-through the main features Attunity Compose I have found interesting.
Compose introduces end-to-end workflow management which includes major stages of data warehouse lifecycle resorting to combination of Inmon (for time variant audible enterprise data warehouse it employs data vault) and Kimball (for presentation layer – star schema) methodologies.
One of the Compose’s benefits is an intuitive user interface in the familiar Attunity Replicate style. This is easy to navigate, access basic commands and start working.
Among the things that can be done with a click of a button are business model creation, mapping staging tables to Data Warehouse and Data Mart objects, creating marts, tracking lineage and monitoring task execution.
The version 3.1 supports a wide range of data sources and data warehouse platforms including:
- Microsoft SQL Server
- Oracle DB
- Oracle Exadata
- Teradata DB
- IBM DB2 LUW
Running on various versions of Windows and in the cloud.
Integration with Replicate gives a power to control and monitor loading tasks and an ease to incorporate new sources into your model or adjust due to a change.
Data profiling and data quality
Data sources can be analysed using basic predefined profiling techniques. There is an option to create pre and post-load scripts to control quality of data by correcting or filtering it out, e.g if you load data into data marts and it does not pass specified criteria then it falls into erroneous mart.
There are several options at your disposal to design the model either using reverse engineering of data sources or loading ERwin model with a possibility of manual adjustments as well as full manual design.
As data models evolve over time Compose automatically tracks and compares data warehouse and data mart physical representations to the model and automatically generates the code.
Compose auto-generates ETL commands for moving data and transforming it, e.g. doing source-target mappings, filtering records. However users can define and add their own SQL based steps in ETL flow and reuse it later.
A developer can easily schedule, execute, monitor tasks and get mail notifications using workflow designer and scheduler. Attunity noted that scheduler can execute tasks in parallel threads and not only in sequential order. Though a task-chaining option is not provided yet and Attunity recommends to use a third party scheduler if needed.
Release management can be done in one click, you will just need to export and import a project to a desired environment.
A definite bonus is an option to auto-generate project documentation and list of business rules to review.
Attunity Compose looks to be a great product with a promising future. I will be definitely going to try it out and see if it lives up to it’s promise. In particular I would like to see how well it functions in areas such as business rules versioning, behavior in case of data warehouse with complex data sources and business logic, ability to handle performance tuning, model relationship management.
We have written many blogs on Attunity Replicate, you can read them here.
OptimalBI is an Attunity partner. Images are via the Attunity.com website.