API Queue Scheduling for SLA-Based Machine Allocation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems face challenges in managing a large number of inconsistent APIs across multiple machines, leading to inefficient utilization of resources and difficulty in determining the minimum number of machines required to meet cumulative Service-Level Agreements (SLAs).

Innovation Solution

A system and method that utilizes a mixed integer programming (MIP) model and heuristics to optimize API scheduling, ranking APIs based on cumulative SLA, and allocating them to computing devices to maximize resource utilization and minimize the number of machines needed to meet SLAs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If APIs are managed in a growing enterprise environment with multiple versions and states, then the number of APIs increases, but the difficulty of managing and locating APIs increases

Engineering Contradiction:

ImproveAPI lifecycle management capabilityVSAvoidAPI management complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the API management system into distinct components: an API catalog for metadata storage, a scheduling system for task allocation, and a resource management layer. This segmentation allows the system to handle large numbers of APIs across multiple versions and states without increasing overall management complexity, as each component handles specific aspects independently.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces an API catalog as an intermediary layer between API consumers and the actual API implementations. This catalog stores metadata, versions, and state information, allowing consumers to locate and manage APIs without directly interacting with the complex backend infrastructure, thus reducing management difficulty despite API proliferation.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If contemporary optimization methods are used for API scheduling, then scheduling can be performed, but machine utilization cannot be maximized for SLA

Engineering Contradiction:

ImproveAPI scheduling capabilityVSAvoidmachine utilization efficiency

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The patent transforms the scheduling problem by changing the optimization parameters from simple task allocation to a multi-objective optimization that includes machine utilization rates, SLA compliance metrics, and cumulative SLA calculations. This allows the system to maximize machine utilization while meeting service level agreements, rather than merely performing basic scheduling.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent implements a feedback mechanism where the scheduling system continuously monitors machine utilization, SLA compliance, and API execution status. This feedback is used to dynamically adjust scheduling decisions, ensuring that machine resources are optimized for SLA adherence while maintaining high productivity in API execution.

Inventive Principle:

Principle #23Feedback

3Reliability

If more machines are allocated to schedule APIs, then SLA compliance can be maintained, but the number of machines required is difficult to minimize

Engineering Contradiction:

ImproveSLA complianceVSAvoidnumber of machines

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent implements dynamic machine allocation where the number of machines assigned to API scheduling is not fixed but adjusts based on real-time workload, SLA requirements, and resource availability. This dynamic approach allows the system to maintain SLA compliance with the minimum necessary machines, rather than over-provisioning statically.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent performs preliminary calculations to determine the minimum number of machines required to meet cumulative SLA requirements before actual API scheduling begins. This preliminary action involves analyzing API metadata, execution patterns, and SLA constraints to pre-determine optimal resource allocation, minimizing the number of machines needed while ensuring reliability.

Inventive Principle:

Principle #10Preliminary action

4Adaptability or versatility

If APIs are deployed across multiple landscapes (development, test, productive), then complete lifecycle management is achieved, but locating and consuming APIs from different teams becomes difficult

Engineering Contradiction:

ImproveAPI lifecycle state managementVSAvoidAPI location and consumption

Core Design Contradiction:

Adaptability or versatilityVSEase of operation

Solution Approach 1:

The patent creates a universal API catalog that serves multiple functions: storing metadata for APIs across all lifecycle landscapes (development, test, productive), providing a unified search interface, and enabling consistent consumption patterns regardless of the API's current state or originating team. This multi-functional catalog simplifies API location and consumption while maintaining comprehensive lifecycle management.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12619484B2System and method of application programming interface scheduling

Publication Date: 2026.05.05 JIO PLATFORMS LTD

AI summary

The present invention provides a robust and effective solution to an entity or an organization by enabling maximization of the utilization of machine resources by optimally allocating the tasks such as application programming interfaces (APIS) in the queue using a set of predetermined instructions. The method further enables finding the number of machines in order to fulfil a cumulative service-level agreement (SLA) of the APIs in the queue using heuristics and the set of predetermined instructions.