40 DynamoDB Interview Questions and Answers

DynamoDB is a fully managed NoSQL database service provided by AWS. It offers fast and predictable performance with seamless scalability. DynamoDB is commonly used for applications that require consistent, single-digit millisecond latency at any scale, such as gaming, IoT, mobile apps, and real-time analytics.

DynamoDB automatically replicates data across multiple Availability Zones within an AWS region, providing built-in high availability and durability. This means your data remains safe and accessible even if a server or an entire data center fails.

In DynamoDB, a partition key is a unique attribute used to distribute data across partitions for scalability. A sort key is an optional second attribute that allows multiple items with the same partition key to be sorted and queried efficiently.

A DynamoDB table is a collection of items, where each item is a set of attributes. Each table must have a primary key, which can be either a single partition key or a combination of partition key and sort key.

DynamoDB automatically scales throughput capacity to meet workload demands using on-demand or provisioned capacity modes. It can handle sudden increases in traffic without manual intervention.

Secondary indexes, such as Global Secondary Indexes (GSI) and Local Secondary Indexes (LSI), allow you to query the table using non-primary key attributes, enabling more flexible and efficient queries.

By default, DynamoDB provides eventual consistency for read operations, meaning data may not be immediately consistent across all replicas. However, you can request strongly consistent reads to ensure you always get the latest data.

Provisioned mode lets you specify the number of reads and writes per second, while on-demand mode automatically adjusts capacity based on traffic. On-demand is ideal for unpredictable workloads, while provisioned is cost-effective for steady workloads.

DynamoDB Streams captures a time-ordered sequence of changes to items in a table, allowing you to build event-driven applications, replicate data, or trigger AWS Lambda functions in response to data modifications.

DynamoDB integrates with AWS Identity and Access Management (IAM) to control access at the table and item level. It also supports encryption at rest and in transit to protect sensitive data.

GSIs allow you to query on non-primary key attributes and can be created or deleted at any time, while LSIs must be defined at table creation and share the same partition key as the base table but have a different sort key. GSIs support eventual consistency, whereas LSIs can support both eventual and strong consistency.

Conditional writes in DynamoDB let you specify conditions that must be met for a write operation (Put, Update, Delete) to succeed. This is useful for implementing optimistic concurrency control, preventing overwrites, and enforcing business rules such as unique constraints.

DynamoDB Streams capture changes to table items in real time. By integrating Streams with AWS Lambda, you can trigger serverless functions in response to data modifications, enabling use cases like real-time analytics, notifications, and data replication.

Hot partitions occur when too many requests target the same partition key, leading to throttling and performance issues. To avoid hot partitions, use high-cardinality partition keys, distribute workload evenly, and consider using randomization techniques or composite keys.

DynamoDB is optimized for denormalized, schema-less data models. For one-to-many relationships, use composite primary keys or secondary indexes. For many-to-many, use adjacency lists or mapping tables. Always design your model based on access patterns to optimize performance.

DynamoDB transactions allow you to group multiple Put, Update, Delete, and ConditionCheck operations into a single, all-or-nothing request. Transactions ensure atomicity and consistency by using a two-phase commit protocol, making them suitable for complex business logic.

DynamoDB has a maximum item size of 400 KB. For larger data, you can store metadata in DynamoDB and the actual content in Amazon S3, linking them via pointers. This pattern is known as the 'S3+Metadata' approach.

Query operations retrieve items based on primary key values and are efficient, while Scan operations examine every item in the table, which can be slow and costly for large datasets. Always prefer Query over Scan when possible.

TTL allows you to define an expiration time for items in a table. DynamoDB automatically deletes expired items, helping manage storage costs and data lifecycle without manual intervention.

DynamoDB provides on-demand and continuous backups (point-in-time recovery) to protect data. You can restore tables to any point within the retention period. For disaster recovery, you can replicate tables across regions using DynamoDB Global Tables.

DynamoDB Global Tables provide fully managed, multi-region, and multi-master database replication. They allow you to deploy a single table across multiple AWS regions, enabling low-latency data access and disaster recovery. Updates in any region are automatically propagated to other regions, supporting active-active workloads.

DynamoDB Global Tables use a 'last writer wins' conflict resolution strategy based on timestamps. If the same item is updated in different regions at the same time, the update with the latest timestamp is retained, ensuring eventual consistency across all replicas.

Adaptive capacity automatically shifts unused throughput from underutilized partitions to 'hot' partitions experiencing higher traffic. This helps prevent request throttling and optimizes resource usage without manual intervention.

Fine-grained access control can be achieved using AWS IAM policies with condition expressions that reference DynamoDB item attributes. This allows you to restrict access to specific items or attributes based on user identity or request context.

Migrating from a relational database to DynamoDB involves denormalizing data, identifying access patterns, designing primary keys and secondary indexes, and possibly using tools like AWS Database Migration Service. It's important to model data for efficient queries rather than traditional normalization.

DAX is an in-memory caching service for DynamoDB that delivers microsecond response times for read-heavy workloads. It is ideal for applications requiring high throughput and low latency, such as gaming leaderboards or real-time analytics.

Best practices include projecting only necessary attributes, monitoring index usage, avoiding excessive indexes, and designing indexes based on query patterns. Regularly review and delete unused indexes to control costs.

Optimistic locking is implemented using a version number attribute and conditional writes. Each update checks the version number to ensure no concurrent modifications have occurred, preventing lost updates and ensuring data integrity.

DynamoDB Streams can trigger Lambda functions on data changes, enabling serverless event processing, data transformation, notifications, and integration with other AWS services without managing servers.

DynamoDB transactions are limited to 25 items or 4 MB of data per transaction. To mitigate these limits, break large operations into multiple transactions, use batching, or redesign workflows to minimize transactional requirements.

Use AWS CloudWatch metrics, DynamoDB's built-in dashboards, and the 'Contributor Insights' feature to monitor throughput, latency, throttling, and hot partitions. Enable detailed logging and use the 'Explain' API for query analysis.

DynamoDB is a managed AWS service and does not natively support hybrid or multi-cloud deployments. However, you can use the DynamoDB API from on-premises or other clouds, or replicate data using AWS Glue, Data Pipeline, or custom ETL solutions.

Many-to-many relationships are modeled using mapping tables or adjacency lists. This approach enables efficient lookups but can increase data duplication and complexity. Careful design is needed to optimize for query patterns and scalability.

Enable encryption at rest and in transit, use IAM roles and policies for least-privilege access, enable VPC endpoints for private connectivity, and audit access using AWS CloudTrail. Consider client-side encryption for highly sensitive data.

DynamoDB is schema-less, allowing you to add or remove attributes without downtime. For backward compatibility, use versioning, default values, and avoid removing attributes that are still in use by existing applications.

Larger item sizes increase read/write costs and latency. Projecting only required attributes in secondary indexes reduces storage and improves query performance. Always minimize item size and index projections for efficiency.

Use DynamoDB Streams to capture data changes and process them with AWS Lambda, Kinesis Data Streams, or Firehose. This enables real-time analytics, dashboards, and integration with data lakes or warehouses.

Use composite keys with time-based attributes, partition data by time intervals, and archive old data using TTL or periodic exports. Design for efficient range queries and avoid hot partitions by distributing writes.

Pricing is based on read/write throughput, storage, and optional features like DAX or Streams. Optimize costs by choosing the right capacity mode, minimizing item size, deleting unused indexes, and using TTL to remove stale data.

DynamoDB is not ideal for complex relational queries, multi-table joins, or transactional workloads exceeding its transaction limits. For such use cases, a relational database like Amazon RDS or Aurora may be more appropriate.