Performance Testing

NOIV's built-in performance testing capabilities help you understand how your APIs perform under load with detailed metrics and analysis.

Basic Benchmarking

Test any endpoint with the benchmark command:

bash

noiv benchmark https://api.example.com/endpoint

This runs 100 requests with 10 concurrent connections by default.

Customizing Load Parameters

Request Count

bash

# Test with 1000 requests
noiv benchmark https://api.example.com/endpoint -n 1000

Concurrency Level

bash

# 50 concurrent connections
noiv benchmark https://api.example.com/endpoint -c 50

Time-Based Testing

bash

# Run for 60 seconds regardless of request count
noiv benchmark https://api.example.com/endpoint -d 60

Combined Parameters

bash

# 500 requests, 25 concurrent, for maximum 30 seconds
noiv benchmark https://api.example.com/endpoint -n 500 -c 25 -d 30

Understanding Results

Sample Output

╭───────────────────────────────────────────────────╮
│ Benchmarking https://api.example.com/users        │
╰───────────────────────────────────────────────────╯

       Benchmark Results       
┏━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ Metric          ┃ Value     ┃
┡━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ Total Requests  │ 100       │
│ Successful      │ 98        │
│ Failed          │ 2         │
│ Total Time      │ 8.45s     │
│ Requests/sec    │ 11.83     │
│ Avg Response    │ 421.33ms  │
│ Min Response    │ 89.12ms   │
│ Max Response    │ 1247.89ms │
│ 50th Percentile │ 387.45ms  │
│ 95th Percentile │ 892.11ms  │
│ 99th Percentile │ 1198.76ms │
└─────────────────┴───────────┘

Key Metrics Explained

Throughput Metrics:

Requests/sec: How many requests completed per second
Total Requests: Total number of requests attempted
Successful/Failed: Request success/failure counts

Response Time Metrics:

Avg Response: Mean response time across all requests
Min/Max Response: Fastest and slowest individual responses
Percentiles: Response time distribution analysis

Percentile Analysis:

50th Percentile (Median): Half of requests completed faster
95th Percentile: 95% of requests completed faster
99th Percentile: 99% of requests completed faster

Performance Testing Strategies

Load Testing Patterns

Baseline Performance

bash

# Light load to establish baseline
noiv benchmark https://api.example.com/users -n 50 -c 5

Normal Load

bash

# Typical production traffic simulation
noiv benchmark https://api.example.com/users -n 500 -c 25

Stress Testing

bash

# High load to find breaking points
noiv benchmark https://api.example.com/users -n 2000 -c 100

Endurance Testing

bash

# Sustained load over time
noiv benchmark https://api.example.com/users -d 300 -c 20

Different Endpoint Types

Read Operations (GET)

bash

# Test data retrieval performance
noiv benchmark https://api.example.com/products -n 1000 -c 50
noiv benchmark https://api.example.com/users/123 -n 500 -c 25

Write Operations (POST)

Note: Be careful with POST requests to avoid creating unwanted data.

bash

# Test with read-only POST endpoints (like search)
noiv benchmark https://api.example.com/search -n 200 -c 10

API Gateway/Load Balancer

bash

# Test infrastructure components
noiv benchmark https://api.example.com/health -n 2000 -c 100

Advanced Performance Analysis

Saving Detailed Results

bash

noiv benchmark https://api.example.com/endpoint -n 1000 -c 50
# When prompted: Save detailed results? [y/n]: y

This creates a JSON file with individual request timings:

json

{
  "url": "https://api.example.com/endpoint",
  "timestamp": 1753362789,
  "config": {
    "requests": 1000,
    "concurrency": 50,
    "duration": null
  },
  "summary": {
    "total_requests": 1000,
    "successful": 987,
    "failed": 13,
    "total_time": 42.5,
    "rps": 23.53,
    "avg_response_time": 412.33
  },
  "detailed_results": [
    {
      "status": 200,
      "time": 0.423,
      "success": true
    }
  ]
}

Performance Regression Testing

Compare performance across deployments:

bash

# Before deployment
noiv benchmark https://staging-api.example.com/users -n 500 -c 25

# After deployment  
noiv benchmark https://api.example.com/users -n 500 -c 25

# Compare the results manually or with scripts

Different Network Conditions

Test from different locations or network conditions:

bash

# Local/fast network
noiv benchmark http://localhost:8080/api -n 1000 -c 50

# Remote/slower network
noiv benchmark https://api.example.com/endpoint -n 500 -c 20

Performance Optimization Insights

Interpreting Results

Good Performance Indicators:

Low response time variance (95th percentile close to average)
High success rate (>99%)
Stable throughput across test duration
Reasonable response times for your use case

Performance Issues:

High response time variance (large gap between average and 95th percentile)
Request failures under normal load
Degrading performance as concurrency increases
Timeouts or errors at moderate load levels

Common Bottlenecks

Database Issues

Symptoms: High response times, failures under load
Solutions: Query optimization, connection pooling, caching

Memory Problems

Symptoms: Increasing response times, eventual failures
Solutions: Memory profiling, garbage collection tuning

Network Limitations

Symptoms: High min response times, consistent delays
Solutions: CDN, load balancing, geographic distribution

Rate Limiting

Symptoms: 429 status codes, consistent failure patterns
Solutions: Implement backoff, optimize request patterns

Best Practices

1. Start Small

bash

# Begin with light load
noiv benchmark https://api.example.com/endpoint -n 10 -c 2

# Gradually increase
noiv benchmark https://api.example.com/endpoint -n 100 -c 10
noiv benchmark https://api.example.com/endpoint -n 500 -c 25

2. Test Realistic Scenarios

Use production-like data volumes
Test during typical usage hours
Include authentication headers if required
Test complete user workflows, not just individual endpoints

3. Monitor System Resources

While running NOIV benchmarks, monitor:

CPU usage on API servers
Memory consumption
Database connections
Network bandwidth
Disk I/O

4. Establish Baselines

bash

# Create performance baseline
noiv benchmark https://api.example.com/users -n 500 -c 25

# Document results for future comparison
# Response time: 245ms avg, 12.5 req/sec

5. Test Regularly

Include performance tests in CI/CD
Run before major releases
Monitor performance trends over time
Set up alerts for performance regressions

Integration with Functional Testing

After Functional Tests

bash

# First verify functionality
noiv test run api_tests.yaml

# Then test performance
noiv benchmark https://api.example.com/endpoint -n 200 -c 20

Performance-Aware Test Suites

yaml

# functional_test.yaml with performance expectations
tests:
  - name: Fast Endpoint Test
    url: https://api.example.com/fast-endpoint
    method: GET
    expected_status: 200
    timeout: 1  # Expect response within 1 second
    
  - name: Slower Endpoint Test  
    url: https://api.example.com/complex-calculation
    method: POST
    expected_status: 200
    timeout: 10  # Allow up to 10 seconds

Troubleshooting Performance Tests

Connection Issues

bash

# If you see many failed requests:
# 1. Reduce concurrency
noiv benchmark https://api.example.com/endpoint -n 100 -c 5

# 2. Check if endpoint is accessible
noiv quick https://api.example.com/endpoint

Rate Limiting

bash

# If you see 429 errors, reduce load:
noiv benchmark https://api.example.com/endpoint -n 50 -c 2 -d 60

Timeouts

bash

# For slow endpoints, use longer duration tests:
noiv benchmark https://api.example.com/slow-endpoint -d 120 -c 5

Next Steps

After mastering performance testing:

HTML Reports - Visualize performance data
HTML Reports - Automate performance testing
Quick Start - Optimize your testing strategy
Installation - Advanced testing options

Important

Always coordinate with your operations team before running performance tests against production systems. High load tests can impact real users.

TIP

Performance testing is most valuable when done consistently. Establish baselines and monitor trends rather than focusing on single test results.

Performance Testing ​

Basic Benchmarking ​

Customizing Load Parameters ​

Request Count ​

Concurrency Level ​

Time-Based Testing ​

Combined Parameters ​

Understanding Results ​

Sample Output ​

Key Metrics Explained ​

Performance Testing Strategies ​

Load Testing Patterns ​

Baseline Performance ​

Normal Load ​

Stress Testing ​

Endurance Testing ​

Different Endpoint Types ​

Read Operations (GET) ​

Write Operations (POST) ​

API Gateway/Load Balancer ​

Advanced Performance Analysis ​

Saving Detailed Results ​

Performance Regression Testing ​

Different Network Conditions ​

Performance Optimization Insights ​

Interpreting Results ​

Good Performance Indicators: ​

Performance Issues: ​

Common Bottlenecks ​

Database Issues ​

Memory Problems ​

Network Limitations ​

Rate Limiting ​

Best Practices ​

1. Start Small ​

2. Test Realistic Scenarios ​

3. Monitor System Resources ​

4. Establish Baselines ​

5. Test Regularly ​

Integration with Functional Testing ​

After Functional Tests ​

Performance-Aware Test Suites ​

Troubleshooting Performance Tests ​

Connection Issues ​

Rate Limiting ​

Timeouts ​

Next Steps ​

Performance Testing

Basic Benchmarking

Customizing Load Parameters

Request Count

Concurrency Level

Time-Based Testing

Combined Parameters

Understanding Results

Sample Output

Key Metrics Explained

Performance Testing Strategies

Load Testing Patterns

Baseline Performance

Normal Load

Stress Testing

Endurance Testing

Different Endpoint Types

Read Operations (GET)

Write Operations (POST)

API Gateway/Load Balancer

Advanced Performance Analysis

Saving Detailed Results

Performance Regression Testing

Different Network Conditions

Performance Optimization Insights

Interpreting Results

Good Performance Indicators:

Performance Issues:

Common Bottlenecks

Database Issues

Memory Problems

Network Limitations

Rate Limiting

Best Practices

1. Start Small

2. Test Realistic Scenarios

3. Monitor System Resources

4. Establish Baselines

5. Test Regularly

Integration with Functional Testing

After Functional Tests

Performance-Aware Test Suites

Troubleshooting Performance Tests

Connection Issues

Rate Limiting

Timeouts

Next Steps