analyze

The Montana compression analyzer binary.

A modular and extensible duplex pipeline for L2 batch submission and derivation. This binary compares compression algorithms (Brotli, Zstd, Zlib) for L2 batch data and outputs performance metrics.

Usage

analyze [OPTIONS]

CLI Flags

Required Flags

None - all flags have sensible defaults.

Optional Flags

• -v, --verbose: Verbosity level (can be specified multiple times: -v, -vv, -vvv)
  • No flag (0): WARN level logging
  • -v (1): INFO level logging
  • -vv (2): DEBUG level logging
  • -vvv (3+): TRACE level logging
• -m, --mode <MODE>: Operation mode (default: batch)
  • batch: Batch submission mode - compress and submit L2 blocks
  • derivation: Derivation mode - decompress and derive L2 blocks from compressed batches
  • roundtrip: Roundtrip mode - batch submission followed by derivation with validation
• -i, --input <PATH>: Input JSON file containing transactions to batch submit (default: static/base_mainnet_blocks.json)
• -o, --output <PATH>: Output JSON file for the submitted batch (default: output.json)
• -c, --compression <ALGORITHM>: Compression algorithm to use (default: brotli)
  • brotli: Brotli compression
  • zlib: Zlib (DEFLATE) compression
  • zstd: Zstandard compression
  • all: Run all compression algorithms and compare results (uses best algorithm for submission)

Operation Modes

Batch Mode (default)

Compresses L2 block data and simulates batch submission.

Workflow:

1. Loads source data from input file
2. Encodes L2 blocks into raw batch data
3. Compresses using specified algorithm(s)
4. Outputs compressed batch to output file with metrics

Example:

# Use default brotli compression
analyze --mode batch --input static/base_mainnet_blocks.json --output output.json

# Compare all compression algorithms
analyze --mode batch --compression all

# Use zstd compression with verbose output
analyze -vv --mode batch --compression zstd

Derivation Mode

Decompresses batches from output file (created by batch mode).

Workflow:

1. Reads compressed batches from output file
2. Decompresses using specified algorithm
3. Outputs decompressed batch data and metrics

Example:

# Decompress batches using brotli
analyze --mode derivation --compression brotli --output output.json

# Decompress with debug logging
analyze -vvv --mode derivation --compression zstd

Note: The compression algorithm must match what was used during batch mode.

Roundtrip Mode

Performs batch submission followed by derivation with data integrity validation.

Workflow:

1. Loads source data from input file
2. Encodes and compresses L2 blocks
3. Decompresses the data
4. Validates that decompressed data matches original
5. Outputs validation results and metrics

Example:

# Validate roundtrip with brotli
analyze --mode roundtrip --compression brotli

# Validate with all algorithms (tests each separately)
analyze --mode roundtrip --compression zstd -v

Note: Roundtrip mode does not support --compression all.

Examples

# Basic usage with defaults (batch mode, brotli compression)
analyze

# Batch mode with all algorithms comparison
analyze --mode batch --compression all -v

# Derivation mode to decompress existing output
analyze --mode derivation --output output.json --compression brotli

# Roundtrip validation with verbose logging
analyze --mode roundtrip --compression zstd -vv

# Custom input/output paths
analyze --input /path/to/blocks.json --output /path/to/output.json

# Maximum verbosity for debugging
analyze -vvv --mode batch --compression all

Output Format

The output JSON file contains:

• Batch number
• Compressed batch data (hex-encoded)
• L1 block number
• Transaction hash
• Optional blob hash (for EIP-4844 batches)

Environment Variables

• RUST_LOG: Controls logging output (overrides verbosity flags if set)
  • Example: RUST_LOG=debug analyze

Exit Codes

• 0: Success
• 1: Pipeline failure (error logged to stderr)