Stateless Resource Management for Heterogeneous Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large-scale data processing environments, managing resources across multiple, heterogeneously managed computing and storage systems with different naming conventions poses challenges for comprehensive resource management, particularly in cloud and hybrid cloud provisioning contexts where resource inter-dependencies and naming conventions differ.
Innovation Solution
A centralized, stateless resource management system employs resource tagging to enable efficient querying and management of resources across multiple naming domains by associating physical and virtual resources with tags that include local and remote identifiers, facilitating the generation of system configuration maps and resource queries to coordinate resource allocation and provisioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If centralized resource management is implemented across heterogeneous systems with different naming conventions, then comprehensive resource management capability is improved, but system complexity increases
Solution Approach 1:
The patent introduces a resource manager that acts as an intermediary between heterogeneous resource systems. This mediator collects resource information from multiple naming domains (e.g., cloud resources, on-premises resources) and translates them into a unified management interface, enabling comprehensive resource management without requiring changes to the underlying heterogeneous systems.
Solution Approach 2:
The resource manager implements a universal approach by supporting multiple resource types and naming conventions within a single management framework. It provides a unified interface that can manage diverse resources (compute, storage, network) from different domains, making the system adaptable to various heterogeneous environments without requiring separate management mechanisms for each resource type.
2Loss of information
If resource information is collected from multiple heterogeneous systems, then comprehensive resource visibility is improved, but information management complexity increases
Solution Approach 1:
The resource manager serves as an intermediary that collects resource information from multiple heterogeneous systems and consolidates it into a unified view. It handles the complexity of information collection by abstracting the details of different naming domains and presenting a consistent interface, thus maintaining information completeness while managing complexity centrally.
Solution Approach 2:
The system segments the information collection process by organizing resources into distinct naming domains (e.g., cloud resources, on-premises resources) while maintaining a unified management structure. This segmentation allows complex information from multiple sources to be systematically organized and managed without overwhelming the management system.
3Reliability
If stateful resource management is used to maintain system state, then resource management reliability is improved, but system performance and scalability deteriorate
Solution Approach 1:
The patent implements a stateless resource management approach where the resource manager does not maintain client state or system state in memory. Instead, it relies on request-specific parameters and external state stores (such as databases or configuration files) to maintain necessary information. This self-service approach eliminates the performance overhead of state maintenance while preserving reliability through proper state management mechanisms.
Data Source
AI summary
A method for managing a resource configuration of a heterogeneously managed computer system. During processing of a resource management request, retrieving a set of resource tags are retrieved. Each resource tag associates a local resource identifier that is assigned within a first configuration management domain with an index identifier that is assigned within a second configuration management naming domain. The retrieved tags are mutually associated utilizing the index identifiers to generating a system configuration map. The resource management request is compared with the system configuration map to identify target resources and corresponding target resource information. One or more resource queries are generated based on the identified target resource and corresponding target resource information.