Examkin

Soru

Zorluk: ZorSelecting and Designing Storage and Database Resources

A global fleet management company is designing a Google Cloud architecture to handle two core workloads for its telematics platform:

1. Real-time ingestion and low-latency point-in-time querying of high-throughput time-series telemetry data originating from millions of vehicle GPS sensors.
2. Archival of unstructured driver inspection PDF logs and vehicle camera images that are accessed frequently for the first 30 days and must automatically transition to low-cost archival storage for long-term compliance retention.

Which TWO storage services and features should the lead cloud architect select to satisfy these requirements while minimizing operational overhead?

  1. Cloud Bigtable to ingest high-throughput time-series vehicle telemetry data with millisecond read/write latency.Cevap
  2. Cloud Storage buckets combined with Object Lifecycle Management to store unstructured logs and automatically transition aged files to Nearline or Coldline storage.Cevap
  3. C
    Cloud Spanner to store unstructured PDF inspection logs and binary camera images to guarantee global multi-region transactional consistency for binary objects.
  4. D
    Customer-Supplied Encryption Keys (CSEK) managed manually on local key management hardware for Cloud Storage object encryption to satisfy standard GCP managed key compliance requirements.

Cevap

The optimal architecture selects Cloud Bigtable for high-throughput time-series vehicle sensor data and Cloud Storage with Object Lifecycle Management for unstructured log and image archival.
Cloud Bigtable is optimized for massive write throughput and time-series data such as telemetry from vehicle sensors. Cloud Storage paired with Object Lifecycle Management provides a cost-effective solution for unstructured documents and images by automatically downgrading storage classes as files age.

Adım Adım Çözüm

1
Analyze workload 1 requirement (high-throughput time-series IoT data).
Identify that Cloud Bigtable is the native Google Cloud recommendation for scalable, high-throughput time-series telemetry data.
Bigtable scales seamlessly for write-intensive IoT streaming data while maintaining low latency.
2
Analyze workload 2 requirement (unstructured blob files with lifecycle transitions).
Identify Cloud Storage with Object Lifecycle Management rules.
Cloud Storage handles unstructured PDF and image blobs, and lifecycle rules automate transition to colder storage tiers based on creation time.
3
Evaluate distractors against architectural best practices.
Reject Cloud Spanner for binary blob storage and reject CSEK when managed encryption services like CMEK or Google-managed keys suffice.
Spanner is for structured relational data; CSEK adds unnecessary administrative burden for standard KMS compliance.

Anahtar Kavram

Matching workload access patterns (high-throughput time-series telemetry vs unstructured blob lifecycle archival) to managed GCP storage resources.
Tahmini Süre:2m 0s
Bu soruyu puanla

Konular

Selecting and Designing Storage and Database Resources