Examkin

Soru

Zorluk: OrtaHigh-Performing Database Solutions

A collaborative document editing platform stores document metadata and change logs in Amazon DynamoDB. The platform experiences write bottlenecks during active collaboration sessions, causing throttling, while users also experience latency greater than 5050 milliseconds when retrieving the latest document updates. The database currently uses a low-cardinality attribute, the project creation date, as the partition key. Which two actions should the solutions architect take to resolve these performance issues? (Select two.)

  1. Change the partition key to a high-cardinality attribute, such as a unique document UUID, to distribute write traffic evenly across partitions.Cevap
  2. B
    Use the project creation date as the partition key and a sequential timestamp as the sort key to order the write operations.
  3. C
    Migrate the workload to Amazon RDS and configure a Read Replica to automatically accept write requests during peak write times.
  4. Deploy an Amazon DynamoDB Accelerator (DAX) cluster to cache read responses and reduce retrieval latency to microseconds.Cevap
  5. E
    Configure the DynamoDB table to use provisioned capacity mode with a fixed high value of Write Capacity Units (WCUs) to handle the spikes, while disabling auto-scaling.

Cevap

Changing the partition key to a high-cardinality attribute like a unique document UUID, and deploying an Amazon DynamoDB Accelerator (DAX) cluster to cache read responses.
The correct options recommend changing the partition key to a high-cardinality attribute like a unique document UUID and deploying Amazon DynamoDB Accelerator (DAX). Changing the partition key distributes the write workload evenly across physical partitions, eliminating the partition-level bottlenecks that cause throttling. Deploying DAX introduces an in-memory caching layer that handles frequent read requests, reducing retrieval latency from milliseconds to microseconds and offloading read traffic from the database.

Adım Adım Çözüm

1
Analyze the database configuration and partition key design to identify the cause of the write bottlenecks.
The use of a low-cardinality partition key (project creation date) results in all write operations for documents created on the same day hitting the same physical partition, causing a hot partition.
DynamoDB distributes data across partitions based on the partition key value. Low-cardinality keys lead to uneven traffic distribution and throttling.
2
Identify the correct design to resolve the write throughput limitations.
Choosing a high-cardinality attribute, such as a unique document UUID, distributes the write operations evenly across all available physical partitions.
High-cardinality partition keys ensure even distribution of reads and writes, maximizing the utilization of provisioned throughput.
3
Determine the appropriate caching solution to reduce read latency below the millisecond threshold.
Deploying Amazon DynamoDB Accelerator (DAX) provides an in-memory cache that intercepts read requests.
DAX reduces read latency from milliseconds to microseconds, offloads the read load from the main DynamoDB table, and does not require modifying application query logic.

Anahtar Kavram

High-performing DynamoDB database design relies on choosing high-cardinality partition keys to prevent hot partitions, and using in-memory caching layers like DAX to achieve microsecond read latency.
Tahmini Süre:2m 0s
Bu soruyu puanla

Konular

High-Performing Database Solutions

Bu konudan daha fazla soru