What is a crucial consideration when recommending a load testing strategy for a newly deployed Tableau Server environment?
A. Testing with the maximum number of users simultaneously to assess the peak performance capacity
B. Focusing solely on the load time of the most complex dashboards available on the server
C. Conducting tests only during off-peak hours to minimize the impact on regular users
D. Limiting the testing to only a few selected reports to reduce the load on the server
A. Testing with the maximum number of users simultaneously to assess the peak performance capacity
Explanation:
Why A is Correct?
Load testing with maximum concurrent users is essential to:
Identify performance bottlenecks (e.g., CPU, memory, or network limits).
Validate scalability under peak demand (e.g., can the server handle 1,000+ users?).
Ensure stability before real users encounter failures.
Tableau’s Performance Testing Guide recommends simulating realistic peak loads.
Why Other Options Are Inadequate?
B. Focusing only on complex dashboards: Ignores system-wide performance (e.g., login storms, extract refreshes).
C. Testing only off-peak: Misses real-world stress scenarios.
D. Limiting to a few reports: Doesn’t reflect typical usage patterns.
Best Practices for Load Testing:
Ramp-up gradually: Start with 100 users, increase to max capacity.
Mix activities: Dashboard views, publishes, extracts.
Monitor metrics: Response times, error rates, resource usage.
Reference:
NIST SP 800-146: Stresses testing under "worst-case" loads.
Final Note:
A is the only strategy that ensures readiness for production. Options B-D risk undiscovered failures during actual peak usage