Examkin

Soru

Zorluk: ZorPerformance and Scalability Optimization

A logistics company is preparing to launch a real-time package tracking portal for a global shopping festival. The portal expects a sudden surge of up to 12 million12\text{ million} tracking requests per minute from users during peak hours. Read queries checking package status account for 98%98\% of the database traffic. The architecture consists of an Application Load Balancer (ALB) routing requests to an Amazon Elastic Container Service (ECS) cluster running on AWS Fargate, backed by an Amazon RDS for MySQL database. Load testing reveals that the database CPU reaches 100%100\% utilization and the ALB drops incoming connections under the sudden surge of traffic.

Which two actions should the solutions architect take to optimize the performance and scalability of the architecture? (Select two.)

  1. Submit a support case to AWS to pre-warm the Application Load Balancer to the expected peak request rate.Cevap
  2. Migrate the database to an Amazon Aurora MySQL DB cluster, deploy Aurora Replicas, and direct the read queries to the reader endpoint.Cevap
  3. C
    Configure the Application Load Balancer to automatically scale dynamically in response to the sudden connection spike without pre-warming.
  4. D
    Enable Multi-AZ deployment on the RDS for MySQL database and route the read queries to the standby DB instance.
  5. E
    Deploy an Amazon ElastiCache for Memcached cluster in a Multi-AZ replication configuration to cache package status.

Cevap

Submit a support case to AWS to pre-warm the Application Load Balancer to the expected peak request rate, and migrate the database to an Amazon Aurora MySQL DB cluster, deploying Aurora Replicas to handle read queries via the reader endpoint.
Pre-warming the Application Load Balancer prepares the load balancer to handle sudden connection spikes without dropping requests. Migrating the database to Amazon Aurora MySQL and using Aurora Replicas allows the application to scale read operations horizontally by offloading read queries to the reader endpoint.

Adım Adım Çözüm

1
Address the load balancer bottleneck for sudden traffic surges.
Determine that the Application Load Balancer must be pre-warmed by AWS Support to handle the immediate peak rate of 12 million12\text{ million} requests per minute without dropping connections.
Dynamic scaling of the load balancer is too slow to react to near-instantaneous bursts of this scale.
2
Scale the database tier to handle the read-heavy workload.
Select Amazon Aurora MySQL with read replicas to distribute the 98%98\% read traffic horizontally.
Aurora read replicas can serve read traffic via the reader endpoint, unlike Amazon RDS Multi-AZ standby instances which do not accept read traffic.
3
Evaluate caching options for high availability.
Reject ElastiCache for Memcached for Multi-AZ replication requirements.
Memcached does not support Multi-AZ replication or clustering for failover; Redis is the correct choice if caching is implemented with Multi-AZ.

Anahtar Kavram

Handling sudden massive spikes in network connection and database read loads by pre-warming application load balancers and horizontally scaling read-heavy relational databases using replicas.
Bu soruyu puanla

Konular

Performance and Scalability Optimization

Bu konudan daha fazla soru