Examkin

Soru

Zorluk: OrtaCost-Optimized Database Design and Capacity Planning

A company is designing the database architecture for a web application that manages ticket sales. The database requirements are split into two categories:

* Ticket transaction records: The write workload is highly unpredictable, experiencing extreme surges up to 15,00015,000 writes per second during flash sales, but dropping to near zero for most of the week.
* User profile sessions: The read workload is highly predictable with a stable baseline of 2,0002,000 reads per second throughout the day. Low latency under 10 ms10\text{ ms} is required.

Which TWO strategies should a solutions architect recommend to design a highly cost-optimized database tier that meets these requirements? (Select TWO.)

  1. Store ticket transaction records in an Amazon DynamoDB table configured in On-Demand capacity mode.Cevap
  2. Store user profile sessions in an Amazon DynamoDB table configured in Provisioned capacity mode with Auto Scaling enabled.Cevap
  3. C
    Store ticket transaction records in an Amazon DynamoDB table configured in Provisioned capacity mode with a static capacity of 15,00015,000 write capacity units.
  4. D
    Store ticket transaction records in an Amazon RDS for PostgreSQL Multi-AZ DB cluster, using read replicas as the primary automated disaster recovery failover mechanism.
  5. E
    Store user profile sessions in an Amazon DynamoDB table configured in On-Demand capacity mode to handle the reads.

Cevap

Store ticket transaction records in an Amazon DynamoDB table configured in On-Demand capacity mode, and store user profile sessions in an Amazon DynamoDB table configured in Provisioned capacity mode with Auto Scaling enabled.
The correct strategy involves combining DynamoDB On-Demand capacity mode for the highly unpredictable ticket transaction workloads and Provisioned capacity mode with Auto Scaling for the stable, predictable user profile session workloads. On-Demand mode instantly handles surges up to 15,00015,000 writes without throttling or paying for idle time, which is highly cost-effective for spiky traffic. Provisioned capacity mode with Auto Scaling is more cost-effective for predictable workloads because it provides a lower unit cost for a guaranteed baseline of traffic.

Adım Adım Çözüm

1
Analyze the ticket transaction workload characteristics.
The ticket transaction workload is highly spiky and unpredictable, ranging from zero to 15,00015,000 writes per second.
Understanding the traffic pattern helps determine whether On-Demand or Provisioned capacity mode in DynamoDB is more cost-effective.
2
Select the optimal DynamoDB capacity mode for ticket transaction records.
Amazon DynamoDB On-Demand capacity mode is chosen.
Since the traffic has extreme surges and drops to zero, provisioned capacity would either lead to throttling (if scaled too slowly) or extreme waste (if over-provisioned). On-Demand mode accommodates instant spikes without manual planning and charges only for what is used.
3
Analyze the user profile sessions workload characteristics.
The session workload is highly predictable with a stable baseline of 2,0002,000 reads per second.
Predictable baselines allow for provisioning capacity units to lower costs.
4
Select the optimal DynamoDB capacity mode for user profile sessions.
Amazon DynamoDB Provisioned capacity mode with Auto Scaling is chosen.
For a stable and predictable read volume, provisioned capacity is significantly cheaper per request than On-Demand mode. Auto Scaling can handle any minor deviations cost-effectively.

Anahtar Kavram

Cost-optimized database capacity planning requires matching database capacity modes to workload predictability. Use On-Demand capacity for spiky, unpredictable workloads, and Provisioned capacity with Auto Scaling for predictable baselines.
Bu soruyu puanla

Konular

Cost-Optimized Database Design and Capacity Planning