Delayable Query Scheduling for Database Resource Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current query optimizers focus on finding the fastest execution plans for database queries, which can be resource-intensive and may not be necessary for all queries, potentially disrupting other processes or wasting resources, especially for queries that do not require immediate execution.
Innovation Solution
A method that uses metadata to determine if a query is delayable, allowing the query optimizer to generate a query plan that can be executed at a later time, optimizing resource usage and avoiding resource bottlenecks by scheduling queries based on predefined criteria and system workload.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If the query optimizer searches for the fastest query plan, then query execution speed is improved, but system resource consumption increases
Solution Approach 1:
The patent changes the execution timing parameter of queries from immediate to delayed based on metadata analysis. By modifying when queries are executed rather than how they are executed, the system achieves resource efficiency without sacrificing execution speed when needed. The query optimizer examines metadata to determine if a query can be safely delayed to a lower-utilization period.
Solution Approach 2:
The patent introduces dynamic query scheduling where the execution time of queries is adjusted based on real-time system conditions and metadata characteristics. Instead of a static immediate-execution model, the system dynamically decides whether to execute queries immediately or schedule them for later, creating a flexible resource management approach that adapts to system load.
2Loss of time
If the query optimizer executes queries immediately, then query responsiveness is improved, but system disruption increases
Solution Approach 1:
The patent performs preliminary analysis of query metadata to identify delayable queries before execution. By pre-classifying queries based on their metadata characteristics, the system can proactively schedule non-critical queries for later execution, preventing system disruption before it occurs while maintaining immediate execution for time-sensitive queries.
3Productivity
If the query optimizer optimizes for speed, then query performance is improved, but resource bottlenecks worsen
Solution Approach 1:
The patent implements periodic query execution patterns where delayable queries are scheduled during periods of low system utilization. Instead of continuous high-intensity query processing, the system creates periodic execution waves - immediate execution for critical queries and scheduled execution for non-critical queries during off-peak periods, thereby maintaining productivity while preserving resource availability.
Data Source
AI summary
Method and apparatus for query optimization using metadata. A first query associated with a first metadata is received. Upon receiving the first query, the first metadata is compared to predefined criteria, to determine that the first query is delayable. Upon determining that the first query is delayable, a first query plan is generated for the first query based on the first metadata. Upon generating the first query plan, a first schedule for executing the first query plan is generated. The first execution query plan may be scheduled to be executed at a later point in time based on the first metadata. Finally, the first query plan is executed according to the first schedule.