Spark, Hadoop, and Snowflake for Data Engineering

Duke University via Coursera

Go to Course: https://www.coursera.org/learn/spark-hadoop-snowflake-data-engineering

Introduction

Create scalable data pipelines (Hadoop, Spark, Snowflake, Databricks) for efficient data handling.

Optimize data engineering with clustering and scaling to boost performance and resource use.

Build ML solutions (PySpark, MLFlow) on Databricks for seamless model development and deployment.

Implement DataOps and DevOps practices for continuous integration and deployment (CI/CD) of data-driven applications, including automating processes.

Syllabus

Overview and Introduction to PySpark

In this module, you will learn how to work with different data engineering platforms, such as Hadoop and Spark, and apply their concepts to real-world scenarios. First, you will explore the fundamentals of Hadoop to store and process big data. Next, you will delve into Spark concepts, distributed computing, deferred execution, and Spark SQL. By the end of the week, you will gain hands-on experience with PySpark DataFrames, DataFrame methods, and deferred execution strategies.

Snowflake

In this module, you will explore the Snowflake platform, gaining insights into its architecture and key concepts. Through hands-on practice in the Snowflake Web UI, you'll learn to create tables, manage warehouses, and use the Snowflake Python Connector to interact with tables. By the end of this week, you'll solidify your understanding of Snowflake's architecture and practical applications, emerging with the ability to effectively navigate and leverage the platform for data management and analysis.

Azure Databricks and MLFLow

In this module, you will practice the essential skills for seamlessly managing machine learning workflows using Databricks and MLFlow. First, you will create a Databricks workspace and configure a cluster, setting the stage for efficient data analysis. Next, you will load a sample dataset into the Databricks workspace using the power of PySpark, enabling data manipulation and exploration. Finally, you will install MLFlow either locally or within the Databricks environment, gaining the ability to orchestrate the entire machine learning lifecycle. By the end of this week, you will be able to craft, track, and manage machine learning experiments within Databricks, ensuring precision, reproducibility, and optimal decision-making throughout your data-driven journey.

DataOps and Operations Methodologies

In this module, you will explore the concepts of Kaizen, DevOps, and DataOps and how these methodologies synergistically contribute to efficient and seamless data engineering workflows. Through practical examples, you will learn how Kaizen's continuous improvement philosophy, DevOps' collaborative practices, and DataOps' focus on data quality and integration converge to enhance the development, deployment, and management of data engineering platforms. By the end of this week, you will have the knowledge and perspective needed to optimize data engineering processes and deliver scalable, reliable, and high-quality solutions.