Skip to content

Phase 2 Performance Analysis Report

Use this structure to produce the final written analysis for submission.

1. Program Understanding

  • Briefly explain the provided baseline code.
  • Describe the original thread model and how work is partitioned.

2. Modification Summary

  • Explain the change to sum of cubes of array elements.
  • Confirm array value constraint (integers < 100).
  • Describe how thread counts 1,2,4,6,8 are configured.

3. Measurement Methodology

  • Hardware and software environment.
  • Timing method (System.nanoTime()).
  • Number of runs per configuration and averaging policy.

4. Results

  • Insert main results table.
  • Insert raw trial table (recommended).
  • Insert performance graphs.

5. Performance Discussion

  • Compare speedup trend as thread count increases.
  • Explain the best-performing thread count.
  • Identify where performance starts to decrease.
  • Discuss overhead causes: thread creation, scheduling, synchronization, and CPU core limits.

6. Conclusion

  • Summarize key findings.
  • State whether multithreading improved performance for this workload.
  • Provide one improvement idea for future testing.