Data Analytics with Snowflake

Snowflake is a cloud-native, managed data platform for big data storage and processing that does not require any hardware or software installation, maintenance, or upgrades and can be connected to in a variety of ways. In this course, learn how to set up a Snowflake free trial account, log into it, and navigate the classic Snowflake UI. Next, practice creating and monitoring virtual warehouses and executing DDL operations. Finally, discover how to create databases and tables, load data into tables, instantiate file format objects, and query tables. Upon completion, you'll be able to list features and use cases of Snowflake, differentiate Snowflake editions, and utilize virtual warehouses and databases to separate data and compute.

Scanning an entire table to return output for a single query is computationally intensive, so many technologies rely on partitioning tables to improve query performance for big data. Snowflake implements a form of partitioning known as micro-partitioning where all tables are automatically divided into micro-partitions that users don't need to manage. In this course, explore and execute queries using Snowflake's web interface, Snowsight. Next, learn how to create dashboards in Snowsight and add various charts. Finally, discover how permanent, temporary, and transient tables in Snowflake differ and how such tables can be created, used, and cloned. Upon completion, you'll be able to view database objects in Snowsight, run queries and build dashboards, and differentiate the types of Snowflake tables.

Data in Snowflake goes through a three-phase lifecycle: current data storage, where the data is stored in regular database tables; Time Travel, where you can access data from a specific time; and Fail-safe, which preserves historical data for a configurable retention period. In this course, learn how to use Time Travel to access historical data using query ID filters, relative time offsets, and absolute timestamp values. Next, practice cloning historical copies of tables into current tables and restoring dropped tables with the UNDROP command. Finally, discover how to install SnowSQL, define and use variables, and spool query results to files. Upon completion, you'll be able to utilize Time Travel and Fail-safe in Snowflake, configure SnowSQL, define and use variables, and utilize SnowSQL features.

Snowflake is a modern data cloud solution/ platform that uses an abstraction known as a stage to provide bulk loading of data for analysis, resulting in faster query processing. Begin this course by exploring the internal and external stages in Snowflake. Then, you will load data using Snowflake's classic user interface (UI). Next, examine erroneous data and discover how to query the data and download the results. You will also load data from the stage into an actual table using the SnowSQL command-line interface. Finally, you will create and use table stages, as well as named stages. When you have completed this course, you will be able to differentiate between internal and external stages, list types and features of internal stages, and load data using the classic Snowflake UI.

Snowflake can load data from the Google Cloud Platform (GCP), Microsoft Azure, and Amazon Simple Storage Service (S3) using external stages. Each cloud platform has one or more techniques for integrating with Snowflake. One method of integration that is supported across all three of the major cloud platforms is storage integration objects. Begin this course by loading data from Google Cloud Storage buckets. Create a Snowflake storage integration object and a GCP cloud storage service account for your Snowflake account. Then, load data from Azure Blob Storage using both the storage integration method and the SAS token method. Finally, integrate with Amazon S3 via storage integrations and using an access key and access key ID credentials. When you have completed this course, you will be able to easily integrate Snowflake with Google Cloud Storage, Azure Blob Storage, and Amazon S3.

Snowflake supports the bulk unloading, or export, of batch data using both internal and external stages. Begin this course by unloading data to all three types of internal stages - user, table, and named. Then, unload data to external stages on the Google Cloud Platform, Microsoft Azure, and Amazon Simple Storage Service (S3). Finally, query data directly from staged files, focusing on the syntax and restrictions of querying data in this manner. When you have completed this course, you will be able to unload data from Snowflake to all three types of internal storage, as well as externally to Google Cloud Storage buckets, Amazon S3 bucket, and Azure Blob Storage.

Partitioning data based on a column is a common technique for performance query optimization in many database technologies. Snowflake is a highly-performant, big-data technology that uses its own form of data partitioning called micro-partitioning that stores data in columnar format. In this course, discover the advantages of Snowflake micro-partitions over regular, static partitioning. Next, examine how caching works in Snowflake. Finally, learn how to perform the clustering of Snowflake tables and how clustering helps the performance of filters, point-lookups, and range queries on the clustering key columns. Upon completion, you'll be able to leverage micro-partitioning in Snowflake, differentiate between retrieval optimization caching and local disk caching, and implement clustering with the correct clustering key.

Search optimization service is a Snowflake feature used to create an auxiliary data structure optimized as a search access path to improve selective point lookup query performance. While search optimization is an extremely handy feature, it can also run up substantial storage and compute costs. In this course, learn how to implement search optimizations by measuring the cost of search optimizations for your table and adding the service to it. Next, differentiate between search optimizations and clustering and discover how to implement partitioning for external tables. Finally, practice working with views in Snowflake and differentiate between non-materialized, materialized, and secure views. Upon completion, you'll be able to leverage search optimization in Snowflake, implement external table partitioning, and use and analyze views.

Data is generally processed using a batch or stream methodology depending on how much time between data generation and processing is acceptable. The Snowflake feature Snowpipes processes data in micro-batches which fall in between these two scenarios. In this course, you will cover the implementation of Snowpipes when data is sourced from an internal Snowflake stage. You will kick things off by looking at data ingestion options in Snowflake from a theoretical standpoint, including the differences between bulk data loading and Snowpipes. Then, you get hands-on to set up the infrastructure for data ingestion: an internal stage for CSV data, a destination table for a data load, and a pipe to carry out the load in micro-batches. Next, you will ingest the data into the destination table and explore how this process can be monitored by tracking the pipe status. Finally, you will implement a Snowflake task to trigger a Snowpipe at regular time intervals.

The Snowpipe feature allows Snowflake to input micro-batches of data as it becomes available, generally within minutes of the data being added to a stage and submitted for ingestion. In this course, you will implement the auto-ingestion of CSV files from external Snowflake stages located on the AWS and Azure cloud platforms. You will begin by setting up a continuous data ingestion pipeline where the data source is located in an Azure Storage Container. This pipeline will include several components, such as queues, enterprise applications, and storage integrations as well as the permissions required to get these pieces to talk to one another. You will then implement something similar with an Amazon S3 bucket as the source of data. This set-up will involve AWS services such as IAM roles, SNS topics, as well as Snowflake objects such as notification integrations and pipes.

Data analytics is the systematic, computational analysis of data or statistics used to discover and communicate meaningful patterns. In business, analytics can be used to extract insights for a business strategy or identify new business opportunities. Snowflake is a managed data platform for big data storage, processing, and analytics that allows for common SQL operations and additional operations. In this course, explore various types of Snowflake join operations and data sampling. Next, learn how to use common table expressions (CTEs) and construct queries. Finally, work with functions related to partitioning, windowing, and ranks. Upon completion, you'll be able to use joins, perform row-based and block-based sampling, construct CTEs, and perform windowing and partitioning operations in Snowflake.

Structured data follows a fixed schema, usually does not contain hierarchical information, and is typically stored in a tabular format. Alternatively, semi-structured data does not adhere to a fixed schema, supports hierarchical information, and offers schema flexibility and standardization. The Snowflake platform offers support for structured and semi-structured data. In this course, learn how to load JSON and XML data in Snowflake tables and examine Snowflake's OBJECT, ARRAY, and VARIANT types. Next, explore data validation, how you can handle source data errors in a load process, and how to query JSON structures in Snowflake. Finally, practice loading and querying XML data, including using Snowflake functions. Upon completion, you'll be able to work with JSON and XML data in Snowflake.

Snowflake is a powerful enterprise Software as a Service (SaaS) offering, and like all SaaS technologies, it is particularly important to administer it correctly, from the perspective of both security and access control and of monitoring and attributing resource usage. Otherwise, costs and security vulnerabilities can quickly pile up as users, roles, and accounts increasingly multiply. In this course, discover how to manage the accounts and users within an organization. Then, explore how Snowflake allows users to configure various properties on objects, using parameters. Next, use resource monitors to impose limits on the number of credits consumed by Snowflake and examine what happens if that limit is breached. Learn how to configure single sign-on using Otka and implement federated authentication. Finally, perform data masking and investigate how Snowflake allows you to redact and censor data at the column level using masking polices.

In the traditional approach to data sharing, consumers receive a copy of the provider data, which can allow for the creation of multiple copies of data. Secure Data Sharing in Snowflake enables the secure sharing of specific database objects without data duplication between accounts. In this course, explore the motivations behind Secure Data Sharing and its implementation, as well as how to create and share a Snowflake share. Next, learn how to use Snowflake Marketplace data, the specific permissions required to create and manage shares, and how to set permissions using queries. Finally, practice implementing cross-region data replication to enable Secure Data Sharing across cloud and geographical boundaries. Upon completion, you'll be able to securely share Snowflake objects with other Snowflake users.

Final Exam: The Snowflake Data Platform will test your knowledge and application of the topics presented throughout the The Snowflake Data Platform track of the Skillsoft Aspire Data Infrastructure Journey.