Pipelined Query Execution Early Exit Backward Propagation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In database query execution, preceding operators in a pipeline continue processing despite an early exit by a LimitOp, leading to unnecessary resource consumption and potential errors due to lack of awareness of the early exit, especially in directed acyclic graphs.

Innovation Solution

A database execution engine propagates early exit information backward through the directed acyclic graph to identify and stop unnecessary processing of preceding operators based on their outgoing links and assigned propagation modes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If preceding operators continue processing after an early exit by LimitOp, then the pipeline maintains continuous operation, but unnecessary resource consumption occurs and potential errors arise

Engineering Contradiction:

Improvequery execution efficiencyVSAvoidresource consumption

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system implements feedback by propagating early exit information backward through the directed acyclic graph from the LimitOp to preceding operators. This feedback mechanism allows preceding operators to detect when downstream operators have exited early and adjust their processing accordingly, stopping unnecessary computation while maintaining pipeline integrity.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary action by pre-assigning propagation modes to operators during query plan generation. These propagation modes determine how early exit information should be transmitted through the operator graph, enabling efficient resource management before actual query execution begins.

Inventive Principle:

Principle #10Preliminary action

2Loss of energy

If preceding operators are stopped due to early exit, then resource utilization is optimized, but complex coordination through the directed acyclic graph is required

Engineering Contradiction:

Improveresource consumptionVSAvoidoperator coordination complexity

Core Design Contradiction:

Loss of energyVSDevice complexity

Solution Approach 1:

The system applies local quality by assigning different propagation modes to different operators based on their specific characteristics and positions in the query plan. Each operator's propagation behavior is tailored to its local context, allowing efficient resource management without requiring uniform complex coordination across the entire operator graph.

Inventive Principle:

Principle #3Local quality

3Loss of energy

If early exit information is propagated backward through the directed acyclic graph, then unnecessary processing is prevented, but additional processing overhead is introduced during query execution

Engineering Contradiction:

Improveresource consumptionVSAvoidquery execution time

Core Design Contradiction:

Loss of energyVSLoss of time

Solution Approach 1:

The system performs preliminary action by establishing propagation modes and early exit detection mechanisms during query plan generation, before actual execution begins. This upfront preparation minimizes the overhead during query execution, as the framework is already in place to efficiently propagate early exit information without requiring complex runtime decisions.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12524410B2Handling early exit in a pipelined query execution engine via backward propagation of early exit information

Publication Date: 2026.01.13 SAP SE

AI summary

In some implementations, there is provided receiving a query request including a top k query operator for query plan generation, optimization, and execution; generating a query plan that includes at least one pipeline of a plurality of operators, wherein the at least one pipeline is associated with a directed acyclic graph; detecting in the query plan a first operator in the at least one pipeline that causes an early exit; and in response to the early exit by the first operator during query execution, processing back through the directed acyclic graph to identify at least one preceding operator that should or should not run given the early exit of the first operator.