Module parallel_executor 

Batch-sequential algorithm execution for process mining.

This module provides computation of process mining algorithms, most notably the Directly-Follows Graph (DFG). On all targets (x86_64, arm64, wasm32), batches are processed sequentially using batch-level aggregation for efficiency.

Thread safety

All parallel functions produce deterministic results identical to their sequential counterparts. Partial results from each thread are merged in a well-defined order, making outputs independent of thread scheduling.

Functions

compute_dfg_parallel

Compute a DFG with constant-latency batch processing.

parallel_available

Check whether parallel execution is available.

parallel_discover_dfg

Discover a DFG using batch-sequential computation. Returns JSON string.

parallel_run_algorithms

Run multiple algorithms in parallel. Returns JSON array of results.

run_algorithms_parallel

Run multiple discovery algorithms in parallel and return their JSON results.