Schema-Based Allocation Limits for Adaptive Job Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing resource allocation methods in multi-user computing platforms are inflexible, leading to inefficient resource utilization and potential failure of job requests due to fixed limits, which can be gamed by developers, and require costly monitoring to ensure fair allocation.
Innovation Solution
Dynamic scaling of resource limits based on the data shape and schema-defined scaling factors for job requests, adjusting allocation limits based on the actual subset of data used in the request, preventing gamesmanship and optimizing resource usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fixed allocation limits are imposed on job requests, then resource exhaustion and malicious failures are prevented, but legitimate job requests may fail due to inflexibility
Solution Approach 1:
The patent implements dynamic resource allocation by adjusting allocation limits based on the actual data shape and schema-defined scaling factors. Instead of fixed limits, the system dynamically calculates allocation limits multiplied by scaling factors derived from the query's data characteristics, allowing the allocation to adapt to varying job requirements while preventing resource exhaustion.
Solution Approach 2:
The patent changes the allocation parameter from a fixed value to a dynamic value based on data shape characteristics. The allocation limit is modified by multiplying with a scaling factor that is determined by the schema and the actual data shape, thereby adjusting the resource allocation parameter according to the specific job request characteristics.
2Device complexity
If fixed allocation limits are used, then resource management is simplified, but resource wastage occurs when limits are too conservative
Solution Approach 1:
The system performs self-adjustment by automatically calculating the appropriate allocation limit based on the query's data shape and schema-defined scaling factors. The resource allocation system serves itself by deriving scaling factors from the actual data characteristics without requiring external intervention or complex monitoring, thereby reducing operational complexity while eliminating resource wastage.
3Reliability
If fixed allocation limits are imposed, then monitoring for fair allocation becomes necessary, but costly monitoring is required to prevent gamesmanship
Solution Approach 1:
The patent applies preliminary action by determining the scaling factor before job execution based on the schema and data shape analysis. The allocation limit is pre-adjusted according to the query's characteristics, preventing gamesmanship before it occurs rather than requiring costly monitoring during execution. The schema-defined scaling factors are applied in advance to ensure fair allocation.
4Productivity
If allocation limits are increased to accommodate larger jobs, then more job requests can succeed, but resource exhaustion becomes more likely
Solution Approach 1:
The patent dynamically adjusts the allocation limit parameter based on the actual data shape and schema-defined scaling factors. The allocation limit is scaled proportionally to the job's data characteristics, allowing larger jobs to receive appropriate resources while preventing uniform increase that would cause resource exhaustion. The scaling factor is derived from the schema and data shape, ensuring proportional allocation.
Data Source
AI summary
Methods and systems for schema-based dynamic adjustment to per-job allocation limits, wherein the allocation limits are set as limits in computing resources allocated to processing a job request. A computing platform may determine a scaling factor for a default allocation limit based on fields selected by a query associated with a job request and the data shape of those fields in the data object referenced by the query. The schema to which the data object conforms may include one or more scale factors specified for its defined fields.