The architecture can be divided into different stages as below:
➡ Data Sources
🔹 Streaming data from IoT devices, sensors, gadgets, etc.
🔹 Non-structured data such as Images, audio, video, etc.
🔹 Semi-Structured data such as csv, json, xml, etc.
🔹 Relational databases
➡ Load and Ingest
🔹 Event Hubs takes care of ingesting streaming data
🔹 Azure Data Factory takes care of batch data and can be scheduled or can be triggered based on some events
➡ Store
🔹 Data is loaded into ADLS Gen2 and Azure Synapse analytics based on the requirements
➡ Process
🔹 Stream Analytics handles the Real time data from Event Hubs
🔹 Azure Databricks integrates big data scenarios with traditional data warehouse
🔹 Azure ML and Azure Cognitive Services take care of the activities related to ML models and provide different AI related services based on the data obtained through Azure Databricks
➡ Serve
🔹 Real time Dashboards are created using Power BI for analytics purpose from streaming datasets obtained from Stream Analytics
🔹 Azure Cosmos DB interacts with Azure Databricks and serve the downstream applications
🔹 Power BI handles the dashboards for static data obtained from Azure Synapse Analytics
Credit: Microsoft
إرسال تعليق