AWS DynamoDB and Schema Design
•
7 likes•4,609 views
DynamoDB and Schema Design The document discusses DynamoDB schema design and core concepts. It covers how to model different types of relationships in DynamoDB, using primary keys, secondary indexes, and global secondary indexes. It also discusses techniques for optimizing performance and minimizing costs, such as using projections, sparse indexes, and sharding indexes. The document provides an overview of DynamoDB components and new transactional APIs.
More Related Content
What's hot
What's hot (20)
Similar to AWS DynamoDB and Schema Design
Similar to AWS DynamoDB and Schema Design (20)
More from Amazon Web Services
More from Amazon Web Services (20)
AWS DynamoDB and Schema Design
- 1. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Lee Packham, Solutions Architect 10th December, 2018 AWS Builders’ Day London DynamoDB and Schema Design
- 2. The how, what and why
- 3. Why are we talking about schemas?
- 4. Why are we talking about schemas? “… for maximizing performance and minimizing throughput costs”
- 5. Mindset SQL: normalize until it hurts, de-normalize until it works - The old way NoSQL - you shouldn't start designing your schema until you know the questions it will need to answer NoSQL: aggregate until it hurts, until it works de-aggregate - The new way???
- 6. • Data Redundancy (eliminate) • Data Integrity and Accuracy (ensure) What and How; SQL • Relationships • Normalization • 1NF • … • Integrity Rules
- 7. • Data size • Data shape • Data velocity What and How; NoSQL • Keep related data together • Use sort order • Use GSIs • Distribute queries
- 9. DynamoDB Core Components • DynamoDB • Amazon DynamoDB is a fully managed NoSQL database service that provides fast and predictable performance with seamless scalability. • Tables, Items, Attributes • Primary Key • simple – Partition Key only • composite – Partition Key + Sort Key • Secondary Indexes • Global Secondary Index (GSI) • Local Secondary Index (LSI) • RCU • WCU
- 10. Relationships, indexes and sharding
- 11. 1:1 relationships Users table Partition Key Attributes SSN = 1234567 name = Anna; email = anna@gmx.net; occupation = … SSN = 2345678 name = Jan; email = jan@t-online.de; occupation = …
- 12. 1:N relationships Device-measurements table Partition Key Sort Key Attributes DeviceId = 1 epoch = 5513A97C temp = 30; pressure = 90 DeviceId = 1 epoch = 5513A9DB temp = 29; pressure = 85
- 13. 1:N relationships (non-homogenous) Users table Partition Key Sort Key Attributes SSN = 1234567 info_type = pers eye_color = blue; height: 170 info_type = proff occupation = Developer; skills: { “fixing code”, … SSN = 2345678 info_type = pers eye_color = green; height: 175 info_type = proff occupation = Doctor; skills: { “fixing people”, …
- 14. Indexes; GSI Overloading Users table Partition Key Sort Key Attributes SSN = 1234567 info_type = pers attr1 = blue; attr2: 170 info_type = proff attr1 = Developer; attr2: 100000 SSN = 2345678 info_type = pers attr1 = green; attr2: 175 info_type = proff attr1 = Doctor; attr2: 120000
- 15. N:M relationships users-accounts table accounts-users-gsi Partition Key Sort Key Partition Key Sort Key SSN = 1234567 account_id = 1 account_id = 1 SSN = 1234567 SSN = 1234567 account_id = 3 account_id = 1 SSN = 2345678 SSN = 2345678 account_id = 1 account_id = 2 SSN = 3456789 SSN = 3456789 account_id = 2 account_id = 3 SSN = 1234567
- 16. Indexes; Attribute Projections Tradeoff: cost for provisioned throughput vs cost of storing extra attributes “… for maximizing performance and minimizing throughput costs”
- 19. Indexes; Replica
- 20. Time-Series Data
- 21. re:Invent 2018
- 22. DynamoDB Transactional APIs Any item Purpose built for developers using DynamoDB in any table In region NEW!
- 23. DynamoDB transactions Simplify your code by executing multiple, all-or-nothing actions within and across tables with a single API call. Single API Call NEW!
- 24. Why developers need transactions Transactions help developers maintain data correctness by coordinating actions across multiple items and tables. Simultaneous writes or updates to multiple items. Check multiple conditions while executing actions. Ensure data is consistent across multiple tables. NEW!
- 25. Use cases for transactions Balance financial transactions Process orders Coordinate actions in multi- player games Update metadata in distributed systems NEW!
- 28. DynamoDB on-demand NEW! Features • No capacity planning, provisioning, or reservations– simply make API calls • Pay only for the reads and writes you perform Key Benefits • Eliminates tradeoffs of over- or under- provisioning • Instantly accommodates your workload as traffic ramps up or down
- 29. Recap
- 30. • Data size • Data shape • Data velocity What and How; NoSQL (full circle) • Keep related data together • Use sort order • Use GSIs • Distribute queries
- 31. Mindset Lambda function Notify change Derivative tables Amazon CloudSearch Amazon ElastiCache
- 32. Thank you!