Examkin

Soru

Zorluk: OrtaHigh-Performing Database Solutions

A fleet management system collects real-time telemetry from 20,00020,000 active vehicles. The data is ingested into an Amazon DynamoDB table. The application uses the current date (formatted as `YYYY-MM-DD`) as the partition key. During peak commuting hours, the application experiences frequent write throttling (`ProvisionedThroughputExceededException`), even though the total consumed capacity is well below the table's overall provisioned write capacity. Which design modification should a solutions architect implement to resolve this database performance bottleneck?

  1. A
    Configure the DynamoDB table with a static high provisioned write capacity mode to handle spiky, unpredictable traffic instead of utilizing on-demand capacity mode.
  2. Modify the partition key design to append a random integer suffix to the date, distributing the write operations across multiple physical partitions.Cevap
  3. C
    Change the partition key to a monotonically increasing timestamp to ensure that incoming data is sequentially ordered and indexed.
  4. D
    Migrate the database to Amazon RDS for PostgreSQL in a Multi-AZ deployment, and configure the application to promote the read replicas to primary during peak write spikes.

Cevap

Modify the partition key design to append a random integer suffix to the date, distributing the write operations across multiple physical partitions.
The correct option outlines write sharding. By appending a random integer suffix to the date partition key, the writes are spread across multiple physical partitions. This distributes the high write rate and stays well within individual partition limits.

Adım Adım Çözüm

1
Analyze the cause of the write throttling.
Since all writes for a given day use the same partition key (`YYYY-MM-DD`), they are routed to the same physical partition in DynamoDB. Individual partition throughput is capped at 1,0001,000 Write Capacity Units (WCUs) per second.
This is a classic 'hot partition' problem where total provisioned capacity is sufficient but partition-level limits are exceeded.
2
Select an optimization technique that distributes writes.
Adding a random suffix (e.g., 11 to 1010) to the partition key (creating values like `2026-07-15.1`, `2026-07-15.2`) distributes the writes across up to 10 different partition keys.
This write sharding technique increases the effective write throughput limits for the daily ingestion partition.
3
Validate against alternative database configurations.
Alternative approaches either fail to solve the partitioning bottleneck or introduce incorrect replication architectures.
Ensures the selected partition strategy is the most performant and correct solution.

Anahtar Kavram

Write Sharding / Synthetic Partition Keys in DynamoDB
Bu soruyu puanla

Konular

High-Performing Database Solutions

Bu konudan daha fazla soru