Virtually access, manage, and deliver data in near-real time through an abstraction layer.
In the past few months, Astera Centerprise has undergone major architecture and feature upgrades to cater to a broader range of data management use-cases and customer base. Keeping up with the flow, another significant addition to this platform is Astera Data Virtualization – a tool that creates and manages virtualized views of data as per custom business specifications.
With this new data virtualization feature, users will be able to connect data from heterogeneous sources and create a single source of truth without physically moving data or writing code. This new functionality will provide users near real-time access to data, fast-tracking analytics and report prototyping, and reducing the data-to-insight time.
Here’s an overview of Astera Data Virtualization’s features:
Virtual Data Models (VDM)
Using virtual data models, users can access data from structured (relational databases, Excel, etc.), unstructured (PDFs, text files, etc.), and semi-structured (EDI, XML, REST API, etc.) sources. In addition, Astera Data Virtualization also gives the option to establish entity relationships, define layout options, and enable caching in the abstraction layer. Once created and deployed, a VDM can be used as a single source entity in simple or complex integration flows.
Build Entity Relationships
Using the drag-and-drop feature, users can build and display relationships between source entities to specify the referential integrity of a virtual database table based on common fields. Moreover, the platform also gives the option to create one-to-one or one-to-many relationships according to the user’s unique business requirements.
Multiple Layout Building Options
To ensure layout consistency in virtualized databases, Astera Data Virtualization offers three layout building options:
- Source layout: It fetches the actual layout from the data source.
- Virtual entity layout: It is a virtual representation of an entity in a virtual database after the model is deployed.
- Cache database layout: It represents an entity inside a cache database. However, this layout is only visible on the back-end to developers, if caching is enabled.
Users can modify the attributes of the database table, such as data types, DB type, length of the fields, etc. and create custom fields as well.
Astera Data Virtualization features database caching functionality that enables faster query execution in the virtual database by temporarily copying source records in a cache database. This translates into higher performance, and significant time and costs savings.
Once caching is enabled for an entity, Astera Data Virtualization will create a temporary database on the server to store the source data. When a virtual database entity is queried, the virtual database retrieves data that was last cached.
Here are some of the benefits of this feature:
- Caching utilizes fewer server resources, especially when fetching data from slow-performing data sources.
- Querying on data imported from external sources can be cumbersome due to the unavailability of resources at any given time. Caching can rectify this issue; as it ensures constant availability of a virtual database despite the status of the source system.
- In scenarios where service providers charge per API call or frequent querying is required, caching can help you cut down the cost by reducing the number of calls made for data retrieval. It also saves the system computing time upon every query execution.
Virtual Data Model Deployments Monitor
The virtual data model deployments monitor serves as a centralized window where users can view and manage all the deployed virtual databases in real-time. It gives all the important health indicators, such as:
- The deployment status, deployment log, and deployment server
- Cache status and cache error
- Start, end, and last updated date
The deployment log shows the job progress of all the tasks involved in the deployment of a virtual database. To get an overview of each entity’s caching properties, go through the entity cache details. It shows different parameters, including cache schedule, last cache status, record count, duration, error records, and more.
In short, the virtual data model deployments monitor gives Centerprise users a bird’s eye view of the virtual data models connected to the server and their operational status.
Astera Query Language
Where business users can leverage the code-free, graphical user interface of Astera Data Virtualization to create and deploy virtual data models, the developers can use Astera Query Language (AQL) to utilize the extended capabilities of querying across the virtual layer.
AQL is a native Centerprise language, modeled after the SELECT statement of SQL, that powers the Astera Data Model (ADM) database engine. However, unlike SQL it automates joins and can connect to non-database sources.
AQL statements can be written manually (in the query window) or generated automatically (by using Multi-Table Query in a dataflow). Either way, you can edit the query to retrieve data from virtual databases.
That’s Not All Folks!
The next release will present to you Astera Data Warehouse Builder – an automated enterprise data warehousing solution that aims to speed up the data journey. It features the tools to help users design, implement, and maintain an enterprise data warehouse, ranging from data source modeling to designing destination models and generating flows, in a code-free environment.
Watch this space to find out about the upcoming features of Astera Data Warehouse Builder. To know more about the data virtualization functionality, download the trial version and see Astera Data Virtualization in action.