You need to define Service Level Objectives (SLOs) for a high-traffic multi-region web application. Customers expect the application to always be available and have fast response times. Customers…

Question

Options

• A90 percentile "" 100ms th 95 percentile "" 250ms th
• B90 percentile "" 120ms th 95 percentile "" 275ms th
• C90 percentile "" 150ms th 95 percentile "" 300ms th
• D90 percentile "" 250ms th 95 percentile "" 400ms th

• A
  4% (1)
• B
  7% (2)
• C
  78% (21)
• D
  11% (3)

Explanation

SLOs should have a small buffer above current observed performance, not be set exactly at current measurements. Setting an SLO exactly at observed performance (option B: 120ms / 275ms) means any minor degradation immediately violates the SLO, leaving no room for natural variance. Option A (100ms / 250ms) is tighter than current performance and would already be violated - this is not a realistic target. Option D (250ms / 400ms) is far too loose, does not reflect actual user experience, and wastes error budget. Option C (150ms / 300ms) adds a small, reasonable buffer above current measurements, giving the team room to absorb minor variance while still reflecting the high-quality experience customers already receive. This follows SRE best practice: SLOs should be aspirational but achievable, and slightly above what is currently being delivered.

No community discussion yet for this question.