When developing applications with a NoSQL database like Azure Cosmos DB, some of your earliest decisions can dramatically impact speed, performance, scalability and costs. To help you create the best data modeling, partitioning, and pricing strategies for your workloads, we’ve released three videos that explain these concepts in less than two minutes. You’ll come to understand the impacts and tradeoffs of various approaches, and how to make the right decisions for your workloads.
Data modeling for speed and cost management
At its most basic, data modeling is how you establish the rules for storing and accessing data. It’s an important first step in database and application design because speed and costs – especially for frequent queries – are greatly impacted by your data modeling strategy. In this two-minute video, you’ll discover the key questions to consider in your data model design, the different approaches you can take, and how Azure Cosmos DB change feed (a hero feature you should know about!) can help.
Data partitioning for performance at scale
Once you’ve got your data model sorted, it’s time to turn your attention to your data partitioning strategy. How you write data to servers impacts read and write efficiency and costs, especially as you scale. This is where a well-planned partitioning approach comes in because your partition key is, well, the “key to success”. (Sorry. I had to.) In this two-minute video, you’ll learn the basics of partitioning, and get tips for how to choose the right partition key for your workload.
Pricing options to save time and money
Choosing the right pricing model for your application is about more than your money – it’s about your time. Azure Cosmos DB offers capacity-based standard and autoscale provisioned throughput options, along with a consumption-based serverless offering. When deciding between the three you need to consider traffic patterns, predictability, SLAs, and whether or not you want to manage capacity. This video will explain your options and give tips for selecting the right option for your workload, all in less than two minutes.
Want more?
- Visit the Azure Cosmos DB channel on YouTube for tips and tricks, technical training, and more
- Find documentation on data modeling and partitioning
- Get more info on choosing between provisioned throughput and serverlessÂ
- Learn all about Azure Cosmos DB
0 comments