Short Query Prioritization in Data Processing Services

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional database systems treat all queries equally, leading to long wait times for short queries as they share resources with long queries, causing a mismatch in customer expectations.

Innovation Solution

Implementing a system that allocates an initial fixed amount of resource time to all queries in a fast queue, then gradually downgrades them to slower queues based on resource usage, ensuring short queries are prioritized and completed quickly.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If all queries are treated equally and share resources, then resource utilization is maximized, but short query response time deteriorates due to queue waiting

Engineering Contradiction:

Improveresource utilizationVSAvoidshort query response time

Core Design Contradiction:

Use of energy by moving objectVSLoss of time

Solution Approach 1:

The patent segments the query queue into multiple priority levels (e.g., short query queue and long query queue) based on predicted query execution time. This segmentation allows the system to differentiate between query types and allocate resources accordingly, preventing short queries from being blocked by long queries while maintaining high resource utilization through balanced queue management.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent dynamically changes the scheduling parameter (queue position) based on the predicted execution time parameter. By analyzing query characteristics and predicting execution time, the system adjusts the priority parameter of each query, moving short queries to the front of the queue and long queries to the back, thereby resolving the contradiction between resource utilization and response time.

Inventive Principle:

Principle #35Parameter changes

2Loss of time

If query prioritization is implemented based on predicted execution time, then short query response time is improved, but prediction accuracy deteriorates leading to misclassification

Engineering Contradiction:

Improveshort query response timeVSAvoidquery execution time prediction accuracy

Core Design Contradiction:

Loss of timeVSMeasurement precision

Solution Approach 1:

The patent performs preliminary analysis of query characteristics (such as query length, complexity metrics, and historical data) before the query is executed to predict its execution time. This preliminary action allows the system to classify queries into appropriate priority queues in advance, improving short query response time while using multiple prediction factors to maintain accuracy and reduce misclassification.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS20250156414A1Short query prioritization for data processing service

Publication Date: 2025.05.15 DATABRICKS INC

AI summary

A cluster computing system maintains a first set of queues for short queries and a set second set for longer queries. The first set is allocated a majority of the cluster's processing resources and processes queries on a first in first out basis. The second set is allocated a minority of the cluster's processing resources which are shared among queries in the second set. Accordingly, the system assigns each query to the first set of queues for a fixed amount of resource time. While a query is processing, the system monitors the query's resource time and reassigns the query to the second set of queues if the query has not completed within the allotted amount of resource time. Thus, short queries receive the necessary resources to complete quickly without getting stuck behind longer queries while ensuring that longer queries continue to make progress.