Examkin

Question

Difficulty: HardPlanning Database Options and Persistent Disk Types

A cloud engineering team is designing the storage architecture for two microservices deployed on Google Cloud Compute Engine:

1. An Event Analytics service requiring sub-10 millisecond latency for massive write-heavy time-series and key-value data ingestion.
2. An Order Processing service requiring a relational schema, full ACID transaction support, standard SQL querying, and regional high availability.

Which TWO database configurations should the team select to meet these requirements while adhering to Google Cloud best practices? (Select TWO.)

  1. Deploy Cloud Bigtable for the Event Analytics service to handle high-throughput NoSQL key-value and time-series data.Answer
  2. Deploy Cloud SQL with regional High Availability (HA) enabled for the Order Processing service.Answer
  3. C
    Deploy Cloud Bigtable for the Order Processing service to execute complex relational queries across multiple tables with full ACID compliance.
  4. D
    Mount Local SSD storage attached to Spot VM instances for the Order Processing service primary database files.

Answer

The team should deploy Cloud Bigtable for the Event Analytics service to manage high-throughput time-series data, and deploy Cloud SQL with regional High Availability for the Order Processing service to satisfy relational ACID transaction requirements.
Cloud Bigtable is ideal for high-throughput NoSQL time-series data with low latency, fulfilling the Event Analytics requirement. Cloud SQL with High Availability (HA) provides a managed relational database supporting ACID transactions and regional automatic failover, fulfilling the Order Processing requirement.

Step-by-Step Solution

1
Analyze the workload requirements for the Event Analytics service.
The workload requires sub-10ms latency and high-throughput write capabilities for time-series and key-value data.
Cloud Bigtable is Google Cloud's managed NoSQL database specifically engineered for high-volume time-series and key-value analytical data.
2
Analyze the workload requirements for the Order Processing service.
The workload requires a relational DB engine, ACID compliance, standard SQL, and regional redundancy.
Cloud SQL offers fully managed relational databases with built-in regional HA options featuring automatic primary to standby failover across zones.
3
Evaluate and eliminate incorrect database and compute choices.
Cloud Bigtable lacks relational SQL join features, and Spot VMs with Local SSD risk immediate data loss upon instance preemption.
Persistent relational workloads require durable storage options and non-preemptible VM instances or managed services.

Key Concept

Selecting GCP Database and Block Storage Solutions Based on Access Patterns and Availability Needs
Rate this question

Topics

Planning Database Options and Persistent Disk Types

More questions from this topic