Testing: Add performance and load testing suite

[sfloess]

Overview

Add performance benchmarks and load testing to ensure JNexus performs well under stress and to catch performance regressions.

Current State

• No performance testing
• No load testing
• No benchmarks for regression detection
• Performance characteristics unknown

Testing Score Impact

Current: Testing A- (90/100)
With this: Testing A+ (98/100)

Proposed Tests

1. Benchmark Suite (JMH)

Add Java Microbenchmark Harness for critical paths:

<dependency>
    <groupId>org.openjdk.jmh</groupId>
    <artifactId>jmh-core</artifactId>
    <version>1.37</version>
    <scope>test</scope>
</dependency>
<dependency>
    <groupId>org.openjdk.jmh</groupId>
    <artifactId>jmh-generator-annprocess</artifactId>
    <version>1.37</version>
    <scope>test</scope>
</dependency>

Benchmarks to add:

@BenchmarkMode(Mode.AverageTime)
@OutputTimeUnit(TimeUnit.MILLISECONDS)
public class NexusClientBenchmark {
    
    @Benchmark
    public void listComponents_cached() {
        // Measure cache hit performance
    }
    
    @Benchmark
    public void listComponents_uncached() {
        // Measure cache miss + HTTP roundtrip
    }
    
    @Benchmark
    public void parseJsonResponse_1000components() {
        // Measure JSON parsing overhead
    }
    
    @Benchmark
    public void regexValidation_complexPattern() {
        // Measure ReDoS prevention overhead
    }
}

2. Load Testing

Test with realistic workloads:

Scenario 1: Large Repository

• 10,000 components
• Pagination (10 pages @ 1000 items each)
• Verify memory usage stays <500MB

Scenario 2: Concurrent Operations

• 5 simultaneous list operations
• Verify thread safety
• No deadlocks or race conditions

Scenario 3: Bulk Delete

• Delete 1,000 components
• Verify progress callback performance
• Measure time per component

3. Stress Testing

Cache Stress:

@Test
void testCacheUnder10000Repositories() {
    // Fill cache with 10K repos
    // Verify memory footprint
    // Verify LRU eviction works
}

Connection Stress:

@Test
void testRetryLogicUnderServerStress() {
    // Simulate flaky server (50% failure rate)
    // Verify retry logic holds up
    // Measure total time with backoff
}

4. Memory Profiling

Add heap dump analysis on large operations:

java -Xmx512m -XX:+HeapDumpOnOutOfMemoryError \
  -jar jnexus.jar list large-repo --show-metadata

5. Performance Assertions

Set performance SLAs in tests:

@Test
@Timeout(value = 5, unit = TimeUnit.SECONDS)
void listComponents_shouldCompleteIn5Seconds() {
    // Fail if >5s for 1000 components
}

Acceptance Criteria

Performance Targets:

• Cache hit: <10ms
• Cache miss + HTTP: <500ms (p95)
• JSON parsing 1000 items: <100ms
• Regex validation: <5ms
• Memory: <200MB for 10K components

Load Targets:

• Handle 10K components without OOM
• 5 concurrent operations without deadlock
• Delete 1000 items in <5 minutes

CI Integration

Add to GitHub Actions (optional):

- name: Run Benchmarks
  run: mvn test -Pbenchmark
  
- name: Upload Benchmark Results
  uses: benchmark-action/github-action-benchmark@v1
  with:
    tool: 'jmh'
    output-file-path: target/jmh-result.json

Benefits

• Catch performance regressions in CI
• Establish baseline metrics
• Validate scalability claims
• Prove efficiency (cache hit rate, etc.)

Priority

Medium - Important for production confidence

Related

• Performance: Add metrics and monitoring instrumentation #60: Metrics/monitoring would complement this
• Code Quality: ReDoS prevention creates thread pool per validation - resource inefficient #53: Would measure executor inefficiency impact