nerdexam
Google

PROFESSIONAL-CLOUD-DATABASE-ENGINEER · Question #193

Your company is migrating from an on-premises database to a single-region Spanner instance. The current database supports 40,000 reads each second, and 7,000 writes each second, at 1- KB row sizes at

The correct answer is A. Select a 4-node Spanner instance.. A single Spanner node (1-000-PU) in a regional SSD configuration can handle up to ~22-500 reads/sec and ~3-500 writes/sec for 1-KB rows. To support 40-000 reads/sec and 7-000 writes/sec, you need at least two nodes (2×22-500-=-45-000 reads; 2×3-500-=-7-000 writes). Of the provide

Designing and Planning a Database Solution

Question

Your company is migrating from an on-premises database to a single-region Spanner instance. The current database supports 40,000 reads each second, and 7,000 writes each second, at 1- KB row sizes at peak. You need to determine the most cost-effective size for the Spanner instance to handle the equivalent current workload. What should you do-

Options

  • ASelect a 4-node Spanner instance.
  • BSelect a 6-node Spanner instance.
  • CSelect a 1-node Spanner instance.
  • DRecommend a multi-region Spanner instance.

How the community answered

(36 responses)
  • A
    81% (29)
  • B
    3% (1)
  • C
    6% (2)
  • D
    11% (4)

Explanation

A single Spanner node (1-000-PU) in a regional SSD configuration can handle up to ~22-500 reads/sec and ~3-500 writes/sec for 1-KB rows. To support 40-000 reads/sec and 7-000 writes/sec, you need at least two nodes (2×22-500-=-45-000 reads; 2×3-500-=-7-000 writes). Of the provided options, a 4-node instance is the smallest choice that exceeds both the read and write requirements, making it the most cost-effective available SKU.

Topics

#Google Cloud Spanner#Database Sizing#Performance Tuning#Workload Migration

Community Discussion

No community discussion yet for this question.

Full PROFESSIONAL-CLOUD-DATABASE-ENGINEER Practice