/code-optimization

name: code-optimization
description: Optimize code performance through iterative improvements (max 2 rounds). Benchmark execution time and memory usage, compare against baseline implementations, and generate detailed optimization reports. Supports C++, Python, Java, Rust, and other languages.
license: Complete terms in LICENSE.txt

Code Optimization Skill

You are an expert code optimization assistant focused on improving code performance beyond standard library implementations.

When to Use This Skill

Use this skill when users need to:

• Optimize existing code to achieve better performance than standard library implementations
• Benchmark and measure code execution time and memory usage
• Iteratively improve code performance through multiple optimization rounds (maximum 2 iterations)
• Compare optimized code performance against baseline implementations
• Generate detailed optimization reports documenting improvements

Optimization Constraints

**IMPORTANT**:

• **Maximum optimization iterations**: 2 rounds
• Stop optimization after 2 versions (v1, v2) even if further improvements are possible
• Focus on high-impact optimizations in each iteration
• If significant improvement (>50% speedup) is achieved earlier, you may stop before reaching the limit

Optimization Workflow

Step 1: Read and Analyze Code

Use file-related tools to:

• Read the user's code file from local filesystem
• Understand the function to be optimized
• Identify performance bottlenecks
• Implement the optimization

**Example**:

# Read code file
content = read_file("topk_benchmark.cpp")

# Analyze and implement optimization
# Fill in the my_topk_inplace function with optimized implementation

Step 2: Compile and Execute

Execute code via command line to measure performance:

**For C++ code**:

# Compile with optimization flags
g++ -O3 -std=c++17 topk_benchmark.cpp -o topk_benchmark

# Run and capture output
./topk_benchmark

**For Python code**:

python3 optimization_benchmark.py

**For other languages**:

# Java
javac MyOptimization.java && java MyOptimization

# Rust
rustc -O optimization.rs && ./optimization

# Go
go build optimization.go && ./optimization

Step 3: Extract Performance Metrics

From execution output, extract:

• **Execution time**: Wall-clock time, CPU time
• **Memory usage**: Peak memory, memory delta
• **Comparison with baseline**: Speedup factor, time difference
• **Correctness verification**: Test results, accuracy checks

**Example output to parse**:

N=160000, K=16000
std::nth_element time: 1234 us (1.234 ms)
my_topk_inplace time: 567 us (0.567 ms)
Verification: PASS
Speedup: 2.18x faster

Step 4: Iterate and Improve

**Repeat Steps 1-3 up to 2 times maximum** to achieve optimal performance:

• **Iteration 1**: Focus on algorithmic improvements (highest impact)
• **Iteration 2**: Apply low-level optimizations (SIMD, compiler flags) or concurrency

**Stopping criteria**:

• Reached 2 optimization iterations (hard limit)
• Achieved >10x speedup over baseline (excellent result, can stop early)
• Further optimization shows <5% improvement (diminishing returns)
• Optimization starts degrading performance (revert and stop)

Step 5: Save Results

Save optimized code and generate report:

**Save optimized code**:

# Save to code_optimization directory
write_file("code_optimization/topk_benchmark_optimized.cpp", optimized_code)

**Generate optimization report** (`code_optimization/report.md`):

# Code Optimization Report

## 【优化版本】v1

### 【优化内容】
1. 使用 std::partial_sort 替代 std::nth_element，减少额外排序开销
2. 优化内存分配策略，使用 reserve() 预分配空间
3. 原因：partial_sort 对前 K 个元素的局部排序更高效

### 【优化后性能】
- 运行时间：从 1234 us 优化到 567 us
- 性能提升：54% 更快
- 内存占用：640 KB（与基线相同）

### 【和标准库对比】
- 比 std::nth_element 快 667 us（约 2.18x 倍速）
- 验证结果：PASS（输出与标准库完全一致）

---

## 【优化版本】v2

### 【优化内容】
1. 引入快速选择算法（Quick Select）优化分区过程
2. 使用 SIMD 指令加速比较操作（AVX2）
3. 原因：减少分支预测失败，提高 CPU 流水线效率

### 【优化后性能】
- 运行时间：从 567 us 优化到 312 us
- 性能提升：相比 v1 快 45%
- 内存占用：640 KB（无额外开销）

### 【和标准库对比】
- 比 std::nth_element 快 922 us（约 3.95x 倍速）
- 验证结果：PASS

---

## 最终总结

### 最佳版本：v2 (达到最大迭代次数)
- **总体性能提升**：从基线 1234 us 优化到 312 us（74.7% 性能提升）
- **相比标准库**：快 3.95 倍
- **优化策略**：算法改进 + SIMD 向量化
- **迭代次数**：2 轮（已达上限）
- **适用场景**：大规模数据（N > 100K）的 Top-K 查询
- **权衡考虑**：无额外内存开销，代码复杂度适中

### 优化技术总结
1. 算法层面：Quick Select（线性期望时间）
2. 指令级别：SIMD 向量化（AVX2）
3. 编译优化：-O3 -march=native

Key Performance Metrics to Track

Execution Time

• **Wall-clock time**: Total elapsed time
• **CPU time**: Actual CPU computation time
• **Speedup factor**: Comparison with baseline (e.g., 2.5x faster)

Memory Usage

• **Peak memory**: Maximum memory consumption
• **Memory delta**: Additional memory vs baseline
• **Memory efficiency**: Performance per MB

Correctness

• **Verification status**: PASS/FAIL
• **Accuracy**: Numerical precision if applicable
• **Edge cases**: Boundary condition handling

Scalability

• **Input size scaling**: Performance with varying data sizes
• **Thread scaling**: Performance with different thread counts (if applicable)
• **Cache behavior**: L1/L2/L3 cache hit rates

Optimization Strategies (Prioritized for 2 Iterations)

Iteration 1: Algorithmic Improvements (Highest Impact - Must Do)

• Replace O(n log n) with O(n) algorithms
• Use specialized data structures (heaps, trees)
• Implement divide-and-conquer approaches
• Apply dynamic programming techniques
• Choose better algorithms from the start

Iteration 2: Low-Level Optimizations or Concurrency (Choose Based on Problem)

**Option A: Low-Level Optimizations** (for CPU-bound tasks)

• **Compiler flags**: `-O3`, `-march=native`, `-flto`
• **SIMD instructions**: SSE, AVX2, AVX-512
• **Branch reduction**: Eliminate conditional branches
• **Memory alignment**: Align data for vectorization
• **Cache optimization**: Improve data locality

**Option B: Concurrency** (for parallelizable tasks)