Learning Azure Cosmos DB

Hey everyone!

I’m extremely delighted to announce the launch of my brand new Pluralsight course on Azure Cosmos DB.

This is a 10-module course that targets developers – and that’s you!

The course should be live on Pluralsight in the next few days. If you’re interested in building next-generation global apps for the web and mobile, just join me for about 6 hours, and get up to speed in a heartbeat!

Here’s what the course delivers:

Part 1: Introduction and Overview

We begin by defining NoSQL, explaining big data, and describing the characteristics of a NoSQL Database. Then we introduce Cosmos DB with a high-level discussion around global distribution, server-side partitioning, and its multi-model capabilities (with support for documents, tables, and graphs). We also introduce the local emulator, which lets you develop for Cosmos DB without an internet connection or Azure subscription. This module has several demos showing how to create an Azure Cosmos DB account, how to download and install the local emulator, and how to use the SQL API to create a collection, populate it with documents, and query for documents using Cosmos DB SQL.

Part 2: Tuning for Throughput and Performance

This module explains how to provision throughput in Cosmos DB. We begin by learning about request units, or RUs. With RUs, you don’t worry about hardware concerns, like CPU, memory, or I/O. Instead, Cosmos DB delivers predictable performance by letting you reserve as many RUs per second that your application needs, and will throttle requests if you start to exceed to throughput limits that you reserve. The portal has many customizable charts and graphs that let you monitor the request unit consumption of your database, and you can whiteboard the cost to estimate your throughput needs using the online request unit calculator. The module concludes with an explanation of how Cosmos DB pricing is divided between storage and throughput.

Part 3: Horizontal Partitioning

This module discusses how Cosmos DB achieves elastic scale via server-side horizontal partitioning. We explain how containers are logical resources that Cosmos DB manages by creating and replicating multiple physical partitions to scale both storage and throughput. For best results, we discuss the considerations to keep in mind when choosing a partition key, which groups related items together for queries, and supports updating multiple items in a transaction using stored procedures. The module concludes by showing how to enable cross-partition queries, where Cosmos DB automatically fans out the query to multiple partitions, and aggregates each partition’s results into a single result for the query.

Part 4: Globally Distributing Data

In this module, we explore the geo-replication capabilities in Cosmos DB, also called turnkey global distribution. In addition to local replication within one Azure data center, Cosmos DB can replicate your database to any number of other Azure data centers in regions located throughout the world. This brings your data closer to your users, giving them low latency and high availability wherever they are located. You will see how easy it is to enable global distribution, and configure your application with a failover sequence of preferred data. We wrap up the module by discussing consistency, and explaining the five consistency levels that let you balance latency and availability with stale reads that occur when querying replicas that are not up to date.

Part 5: Data Modeling for the SQL API

Using the SQL API, you model items in the database as JSON documents. This module starts by calling out some of the key differences in data modeling between a traditional relational database platform (like SQL Server) and a JSON document database (like Cosmos DB with the SQL API). We then examine the special properties, including the resource ID and self-link properties that uniquely identify each resource, and the URI factory which simplifies the task of constructing the self-link to any resource in the database. The module concludes by showing how the Data Migration Tool can be used to transform and import data from SQL Server to Cosmos DB.

Part 6: Querying Documents

This module explores the special version of SQL used by Cosmos DB for querying JSON documents in a collection
with the SQL API. We explain how the query language is rooted in JSON and JavaScript semantics, and then
dive into numerous demos that show how to filter, sort, iterate arrays, and perform intra-document joins,
using the many available operators and built-in math, type checking, string, array, aggregate, and spatial
functions provided by Cosmos DB SQL.

Part 7: Programming with the .NET SDK

This module shows how to use the .NET SDK with the SQL API to build client applications for Cosmos DB. We cover how to use the SDK for working with databases and collections, creating and querying documents, indexing, and resource tokens based on users and permissions.

Part 8: Programming the Cosmos DB Server

This module teaches you how to write server-side code that runs within the Cosmos DB service, including stored procedures, triggers, and user-defined functions. We also show you how to cope with throttled requests, and how to implement a continuation model for long-running processes.

Part 9: Using the Table API

This module explains the Table API in Cosmos DB, which provides 100% compatibility with the original Azure Table Storage NoSQL service, but adds predictable reserved throughput, and automatic indexing. We discuss the key-value data model used by the Table API, explain when it makes sense to use this API and when it doesn’t, and show how to migrate an existing Azure Table Storage account and application to Cosmos DB and the Table API.

Part 10: Using the Gremlin API

This module explains the Gremlin API, which provides a graph data model over Cosmos DB. This API is based on the Apache TinkerPop open source project, and defines a set of steps that can be used to populate a graph with vertices (entities) and edges (relationships). After populating the graph, you can run Gremlin queries that traverse the many relationships defined in the database. We’ll create several demos, including a simple graph, airport model, and a multi-model comic book catalog.

I hope this has whet your appetite to learn more about Cosmos DB!

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s

%d bloggers like this: