Soru

Zorluk: ZorSelecting and Designing Storage and Database Resources

A digital health enterprise is designing a multi-region clinical monitoring architecture on Google Cloud. The solution must fulfill two specific operational requirements:

1. Ingest and store real-time time-series telemetry data emitted by millions of wearable patient devices, requiring single-digit millisecond write latency and high-throughput key-value storage.
2. Store core patient administrative and treatment records requiring strong multi-region ACID compliance, high availability across continents, and ANSI SQL query support.

Which TWO database services should you recommend to meet these requirements? (Select TWO.)

  1. Cloud Bigtable to handle high-throughput, low-latency time-series IoT device telemetry data.Cevap
  2. Cloud Spanner to host patient records requiring multi-region relational consistency and ANSI SQL compliance.Cevap
  3. C
    Cloud SQL for PostgreSQL with cross-region read replicas to serve multi-region write transactions for patient records.
  4. D
    Storage Transfer Service to stream real-time telemetry events directly into cold Cloud Storage buckets.

Cevap

The optimal architecture combines Cloud Bigtable for real-time IoT time-series telemetry ingestion and Cloud Spanner for globally consistent, multi-region relational patient records.
Selecting Cloud Bigtable provides the required low latency and massive scale for real-time time-series IoT telemetry. Selecting Cloud Spanner delivers global multi-region transactional consistency with ANSI SQL compatibility for patient record management.

Adım Adım Çözüm

1
Analyze workload 1 (telemetry data)
Identified high-volume time-series IoT data needing single-digit millisecond latency.
Cloud Bigtable provides scalable wide-column NoSQL storage suited for time-series access patterns and high write throughput.
2
Analyze workload 2 (patient records)
Identified relational data requiring multi-region ACID consistency and ANSI SQL support.
Cloud Spanner is the native GCP database designed for globally distributed relational databases requiring multi-region ACID transactions.
3
Evaluate and eliminate unsuitable database alternatives
Cloud SQL cannot support multi-region write scalability; Storage Transfer Service is not an ingestion pipeline engine.
Ensures architectural patterns align with GCP managed service limits and capabilities.

Anahtar Kavram

Selecting GCP Database Services for Hybrid Workloads (NoSQL Time-Series vs Global Relational ACID)
Tahmini Süre:2m 0s
Bu soruyu puanla