- Konzultační služby
Azure Analytics Services: 2-Day Online Workshop
This 2-day Azure Analytics Services workshop contains theoretic and practical tasks. It is also designed to help you to start working with Azure Synapse, Stream Analytics, Analysis Service etc.
Content:
The most common problems in data analysis are data silos, performance constraints, solution complexity and escalating costs. For data analysis in the cloud, different data solutions are created by determining the needs of the customer with different services. Projects stay in the cloud because it offers many different options for every need. It can be adjusted according to your budget and services can be scaled up and down without encountering high costs. You have unlimited resources for storing data. Customers can benefit from Azure Analytics in many projects such as analytics services and recommendation engines, loss analysis, bi reporting, demand forecasting. While doing this, they do not experience any loss of performance.
Day 1: Information and presentation about the content of the following Analytics services
• Synapse Analytics: Azure Synapse Analytics Workspace introduction, Hubs usage content, pipeline creation, and trigger, data loading, data lake organization, workload management, Synapse Analytical Runtimes, Using Linked services, Azure Synapse Data Explorer, Data Transformations with data flow, Transform with spark, with Spark machine learning process, Serverless SQL pools, dedicated SQL pools will be explained. The advantages of the environment where all operations can be done in one place will be stated in Synapse analytics. The options to progress with flows, whether with or without code, will be shared with the customer.
• Stream Analytics: Examples of topics such as data streams, event processing with stream analytics, data streams by azure stream analytics, Stream analytics advantages, creating stream analytics input-output, transforming data, fraud detection of stream analytics, anomaly detection and broadcasting it on power bi will be given.
• Analysis Services: It will be shown how to use complex models in business intelligence, its performance and speed will be specified, tabular and multidimensional models will be explained, on-premises SQL server vs azure analysis service will be compared and their advantages will be mentioned. Analysis service gateway usage will be displayed.
• Data Factory: Data Loading, Data transformation, Pipelines, the advantage of managing the code-free ETL process will be mentioned, and the benefits of performing the data ingestion phase with a data factory will be mentioned.
• Databricks: The ease of use of Databricks with both a data engineer and a data scientist, the benefits of using apache-spark, Azure Databricks, the speed and performance of data processing with the spark engine, and Databricks data reading and code execution will be shown.
Day 2: Synapse Analytics Demos/Labs
• Setting up the Azure Synapse Analytics environment
• Dedicated SQL Pool installation
• Apache Spark Pool installation
• Data acquisition with sample copy data
• Transforming data in data flow
• Display of linked service links
• Showing machine learning in different languages with Spark pool, using notebook
• Using the stored procedure
• Data query examples with SQL script
• Data lake storage gen2 usage
• Sample reporting display
• Easy use of the customer in the environment by making a Power BI connection
In this section, the use of Synapse analytics as a data warehouse, reporting examples, data visualization with Power BI , data science examples with spark are given according to the customer, and it is aimed that the customer will get the best benefit from the service.