Examkin

Soru

Zorluk: OrtaPerformance and Scalability Optimization

A smart utility provider is designing a new serverless telemetry platform to ingest power consumption data from 5,000,0005,000,000 smart meters. Each meter transmits a 2 KB2\text{ KB} JSON payload every 1515 seconds. During grid emergency events, the system must support flash traffic where meters transmit data every second. The application must process the incoming telemetry in near-real-time to detect anomalies, store the data for long-term historical analytics, and support a high-volume dashboard that queries the latest meter status with sub-second latency. Which combination of architectural decisions should a Solutions Architect recommend to achieve optimal performance and scalability? (Select TWO.)

  1. Ingest telemetry data using Amazon Kinesis Data Streams, and configure AWS Lambda with a Parallelization Factor to process the records and write them to an Amazon DynamoDB table.Cevap
  2. Deploy an Amazon Aurora PostgreSQL database with Aurora Replicas, configure Aurora Auto Scaling to dynamically scale the replicas based on CPU utilization, and route dashboard queries to the reader endpoint.Cevap
  3. C
    Ingest telemetry data using an Application Load Balancer (ALB) and an Auto Scaling group of Amazon ECS tasks, relying solely on default ALB target tracking scaling policies to handle the sudden transition to 11-second transmission intervals.
  4. D
    Scale the dashboard query performance during peak loads by deploying an Amazon RDS for PostgreSQL database in a Multi-AZ configuration, and configure the application to send read queries to the passive standby instance in the secondary Availability Zone.
  5. E
    Cache the latest meter statuses and session tokens using an Amazon ElastiCache for Memcached cluster, configuring multi-region replication and data persistence to prevent cache loss during failovers.

Cevap

The optimal scalable design uses Amazon Kinesis Data Streams, AWS Lambda with Parallelization Factor, and Amazon DynamoDB for ingestion, and Amazon Aurora PostgreSQL with reader Auto Scaling and Aurora Replicas for scaling read queries.
Using Amazon Kinesis Data Streams ensures the system can absorb large streams of telemetry data. Processing with AWS Lambda using a Parallelization Factor allows faster stream consumption by running multiple Lambda invocations concurrently for each shard, which accelerates writing to Amazon DynamoDB. To scale database reads, Amazon Aurora PostgreSQL read replicas scale out automatically and are load-balanced via the reader endpoint to handle high-frequency dashboard queries.

Adım Adım Çözüm

1
Select ingestion and processing components designed for real-time high-throughput streams.
Amazon Kinesis Data Streams coupled with AWS Lambda (Parallelization Factor) and Amazon DynamoDB.
Allows massive horizontal scaling of the ingestion layer while handling concurrent record processing on individual shards without increasing shard count.
2
Select database read scaling components for high-volume dashboard queries.
Amazon Aurora PostgreSQL read replicas and reader endpoints with Auto Scaling.
Offloads the primary database engine by routing read traffic to scaled replicas, preventing dashboard query degradation.
3
Exclude solutions relying on Application Load Balancer target tracking without pre-warming for instant spikes, RDS Multi-AZ passive standbys for reads, and Memcached for persistent replication.
Eliminated options using sub-optimal patterns.
These architectures violate performance efficiency and reliability best practices by assuming features that these services do not natively support.

Anahtar Kavram

Decoupling ingestion and scaling read capacity independently are core design strategies for high-performance and high-scalability workloads on AWS.
Tahmini Süre:3m 0s
Bu soruyu puanla

Konular

Performance and Scalability Optimization

Bu konudan daha fazla soru