Data Center

The Data Center concept is a strategic design choice that provides a structured framework for managing the entire data lifecycle.

Data Center: A Centralized Data Hub

The Data Center is a logical, centralized repository for all data-related objects, assets, and functionalities. It acts as the core of the platform's data ecosystem, providing a unified environment for ingesting, preparing, storing, and making data accessible for various analytical and operational applications.

Popular Concepts under the Data Center modules are as defined below:

• Data Connectors: These are the entry points for data. A Data Connector is a pre-built configuration that establishes a secure, persistent connection to an external data source, such as a database, an API, or a file storage system.

• Data Store: A persistent storage unit within the platform. A Data Store serves as a final destination for processed and curated data, providing a structured and optimized repository for long-term storage and use.

• Data Sheet: A specific, tabular view of data, typically residing within a Data Store. A Data Sheet is similar to a table or spreadsheet and is used for specific reporting, dashboards, or as a source for other applications.

• Data Set / Data Service: The source of raw data. A Data Set is a defined collection of data from a source that provides controlled access to that data, ensuring consistent and secure retrieval.

  • Widget: A widget can be instantly generated from a Dataset and subsequently utilized on the homepage to construct customized, insightful visualizations.

• Data Preparation: This refers to the platform's module or tool dedicated to transforming and cleansing data. It enables users to perform enrichment, aggregation, and other transformations on raw data, preparing it for analysis or storage.

• Tables: The fundamental units of structured data. Tables are the most common way to organize data within a Data Store or Data Sheet, consisting of rows and columns. The Table tab displays column names with column types for the ease of users.

• Function (Micro Function): A reusable piece of logic. A Function is a modular component, often implemented with custom code (e.g., Python), that encapsulates a specific transformation, calculation, or business rule to be applied to data.

• Data Sandbox: This refers to the area where users can upload and work with flat files, such as CSV and Excel. It provides a dedicated, temporary space for ingesting and exploring these types of data before they are moved into a more permanent structure.

• Feature Store: A centralized repository for machine learning features. A Feature Store ensures that features used for training and inference are consistent, versioned, and easily accessible to data scientists.

• Data as API: The mechanism for exposing data. Data as API enables a data asset, such as a Data Sheet or Table, to be accessed programmatically via a RESTful API endpoint, allowing for seamless integration with external applications.

• Data Forms: Interactive tools or structures, typically designed as a screen or document, used for the systematic collection of specific information from users.

• Data Quality: Measures the accuracy, completeness, and consistency of your data, making sure that all subsequent analytics, reports, and models built on the platform are reliable and trustworthy.