AtScale Tutorials

**Overview**
AtScale tutorials for Databricks are datasets that serve as the backing data for AtScale sample semantic models. AtScale semantic models are stored in the Semantic Modeling Language (SML) format and they are freely accessible via a GitHub repository.

With these datasets, customers can explore and query industry vertical data models using a variety of business intelligence and data science tools. The AtScale Semantic Layer Platform (Community Edition) is a free download at www.atscale.com.

With AtScale's universal semantic layer. customers can perform advanced analytics using a business-friendly, OLAP interface using the Databrick's Lakehouse and popular tools like Power BI, Tableau, Excel and Looker.

A semantic layer platform delivers business value by acting as an intermediary between complex data sources and end-user queries. It simplifies the access to data through a unified, business-friendly interface, allowing users with varying levels of technical expertise to understand and analyze data without needing to know complex query languages or database structures.

This accessibility leads to better, faster decision-making across all levels of an organization. Moreover, by standardizing definitions and metrics, a semantic layer ensures consistency and accuracy in reporting and analytics, reducing errors and misinterpretations.

**Link to Documentation**
https://github.com/AtScaleInc/sml-models

**Use cases**

**Retailer Model Tutorial**
The Internet Sales tutorial dataset supplies the data for its companion AtScale semantic model available in AtScale's public Github model repository so that users can learn how to create a semantic model on Databricks data for a retail use case.

The Internet Sales dataset is adapted from the Microsoft AdventureWorks tutorial database which simulates a manufacturing company that sells bicycles and bicycle accessories, including product inventory, sales orders, employee information, and customer data. It serves as a comprehensive example for demonstrating data warehousing, star schemas, reporting, data integration, and analytics practices.

This sample dataset and semantic model demonstrate the following modeling concepts:

1. Single fact model
2. Calculated Columns
3. Time-relative calculations
4. Query Dataset
5. Row-level security

**Wholesaler/Distributor Model Tutorial**
The World Wide Importers dataset is a comprehensive example of a wholesaler/distributor, developed primarily for demonstrating SQL Server capabilities. This rich dataset represents a fictional company, World Wide Importers, specializing in the wholesale and distribution of various products globally. It includes detailed records spanning several tables, encompassing sales, purchasing, stock, people, customers, suppliers, and logistics. The dataset is designed to showcase a wide range of queries and data manipulation scenarios, from basic CRUD operations to complex analytical tasks, thus providing a realistic environment for testing, development, training, and benchmarking database tools and techniques.

This model demonstrates the following modeling concepts:

1. Multi-fact model
2. Calculated Columns
3. Time-relative calculations
4. Query Dataset

**Performance Benchmark Model**
The TPC-DS (Transaction Processing Performance Council Decision Support) model dataset is a widely recognized benchmarking standard designed to evaluate the performance of data warehousing and business intelligence systems. It simulates a real-world retail company focusing on sales across multiple channels. The dataset includes various dimensions such as store, item, customer, and promotion, and fact tables like sales, inventory, and returns, reflecting typical business activities and analytical queries. TPC-DS supports a wide array of query types and data volumes, making it an essential tool for assessing the efficiency, scalability, and data processing capabilities of modern data management systems.

**Product details**
Sample tables for World Wide importers sample include:
1. Date
2. Customer
3. Employee
4. Fact_Sale
5. Fact_Purchase
6. Fact_Order

For more details, see the [AtScale Tutorial Models Github Repository](https://github.com/AtScaleInc/sml-models)