Skip to main content

Module distributed

Module distributed 

Source
Expand description

Distributed query execution — plan queries across the swarm.

The distributed planner assigns cells to nodes based on cache locality and available capacity. The coordinator writes a query manifest to storage, executes local tasks, polls for partial results from workers, merges them, and returns the final result.

Structs§

CellInfo
Information about a cell needed for query planning.
NodeInfo
Information about a node from the world view.
PlannedTask
A task assigned to a specific node in a distributed query.
QueryManifest
Query manifest written to storage for distributed execution.

Enums§

NodeState
Node state in the swarm.
QueryPlan
Query execution plan.

Functions§

cleanup_query
Clean up query directory after completion.
create_manifest
Create a new query manifest.
extract_alive_nodes
Extract node information from the world view (helper for runtime). This will be called by the runtime layer that has access to WorldView.
generate_sql_fragment
Generate SQL fragment for a query (simplified for v1).
manifest_path
Query manifest path in storage.
partial_result_path
Partial result path for a node.
plan_query
Plan a query: decide whether to execute locally or distribute across nodes.
read_manifest
Read query manifest from storage.
read_partial_result
Read partial result from Arrow IPC file.
write_manifest
Write query manifest to storage.
write_partial_result
Write partial result as Arrow IPC file.