Examkin

Question

Difficulty: HardHigh-Performing Database Solutions

A financial technology company is launching a new ledger application that records transactions for millions of active accounts. The application requires a database tier that can handle a sustained write volume of 25,00025,000 transactions per second with a write latency of less than 1010 milliseconds. Additionally, the application must support read queries of account balances with a latency of less than 11 millisecond for hot accounts during peak hours. The database must scale automatically to handle sudden spikes in traffic. Which of the following database design and caching configurations should a solutions architect implement to meet these requirements? (Select TWO.)

  1. Design the Amazon DynamoDB table using a high-cardinality attribute, such as `AccountID`, as the partition key to distribute write requests evenly across partitions.Answer
  2. Deploy an Amazon DynamoDB Accelerator (DAX) cluster to cache read requests for hot accounts, reducing read response times to microseconds.Answer
  3. C
    Design the Amazon DynamoDB table using `TransactionDate` as the partition key to store and query transactions chronologically.
  4. D
    Provision an Amazon RDS for PostgreSQL DB instance with a Multi-AZ deployment, and configure a Read Replica to serve as the primary automated failover target during database outages.
  5. E
    Configure the DynamoDB table with Provisioned Capacity Mode using fixed Write Capacity Units (WCUs) and Read Capacity Units (RCUs) calibrated for average load to minimize cost during traffic spikes.

Answer

To meet the requirements, the database must be designed with a high-cardinality partition key like the account identifier in Amazon DynamoDB to distribute writes, and a DynamoDB Accelerator (DAX) cluster must be deployed to provide microsecond read latency for hot accounts.
Designing the DynamoDB table with a high-cardinality partition key ensures writes are distributed evenly, avoiding partition bottlenecks. Deploying a DynamoDB Accelerator (DAX) cluster addresses the microsecond read latency requirement for hot accounts by serving read queries directly from an in-memory cache.

Step-by-Step Solution

1
Analyze the write throughput and scalability requirements.
A sustained write load of 25,00025,000 transactions per second requires a horizontally scalable database. Amazon DynamoDB can handle this workload, provided that data is distributed evenly.
Avoiding hot partitions is critical to preventing write throttling in DynamoDB.
2
Choose the optimal partition key for the DynamoDB table.
Selecting a high-cardinality attribute like the account identifier ensures write requests are evenly spread across multiple physical partitions.
Monotonically increasing keys or low-cardinality keys like dates will concentrate writes on a single partition, causing write bottlenecks.
3
Address the microsecond read latency requirement for hot accounts.
Deploy Amazon DynamoDB Accelerator (DAX) as an in-memory cache in front of the DynamoDB table.
DAX reduces read latency from single-digit milliseconds to microseconds, offloading read requests from the DynamoDB table for hot keys.

Key Concept

Scaling database write throughput via partition key design in Amazon DynamoDB and optimizing read latency for hot keys using DynamoDB Accelerator (DAX).
Rate this question

Topics

High-Performing Database Solutions

More questions from this topic