Management Computer Selecting Integrated Systems By Resource Ratios
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer systems face challenges in efficiently managing resources across integrated systems with varying workloads, leading to inefficient resource redistribution and increased overhead during load changes.
Innovation Solution
A computer system with multiple integrated systems, each with different resource ratios, is managed by a management computer that selects the appropriate system for workload execution based on resource requirements, reducing the need for large-scale resource redistribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If resources are redistributed between integrated systems to handle workload changes, then resource utilization is improved, but redistribution overhead increases
Solution Approach 1:
The system segments resources into fixed allocations within integrated systems rather than allowing continuous redistribution. Each integrated system maintains dedicated resources (CPU, storage, network) in fixed ratios, eliminating the need for dynamic resource movement while maintaining efficient utilization through workload-based system selection.
Solution Approach 2:
The system implements dynamic workload routing based on real-time resource requirements. The management computer dynamically selects which integrated system executes each workload by comparing required resource ratios with available system configurations, achieving adaptive resource utilization without physical resource redistribution.
2Loss of time
If integrated systems are configured with fixed resource ratios, then redistribution overhead is reduced, but adaptability to various workloads decreases
Solution Approach 1:
Different integrated systems are configured with different resource ratio characteristics tailored to specific workload types. For example, some systems have CPU-heavy configurations while others have storage-heavy configurations. This allows each system to be optimized for particular workloads while the collection of systems provides overall adaptability through selective workload routing.
Solution Approach 2:
The management computer provides universal workload management across diverse integrated systems. By maintaining a catalog of resource ratios for each system and implementing intelligent workload matching, the system can handle various workload types (database, web service, batch processing, etc.) across multiple specialized systems, achieving both fixed-configuration efficiency and multi-workload adaptability.
3Productivity
If multiple integrated systems with different resource ratios are deployed, then workload matching is improved, but system complexity increases
Solution Approach 1:
The management computer serves as an intermediary between workloads and integrated systems. It maintains a catalog of resource ratios for each system and implements the matching logic, shielding users from the complexity of multiple system configurations. Users simply submit workload resource requirements, and the management computer handles the complex matching and routing to appropriate systems.
Solution Approach 2:
The system manages complexity by parameterizing resource ratios as standardized metrics. Each integrated system is characterized by specific resource ratio parameters (CPU-to-storage, storage-to-network, etc.), and the management computer uses these parameters for automated matching. This parameterization transforms complex heterogeneous system configurations into manageable quantitative comparisons.
Data Source
AI summary
In a computer system having a plurality of integrated systems of servers and storage apparatuses and having a management computer, each of the plurality of integrated systems has a plurality of resources having different ratios, and the management computer, depending on the ratios of the plurality of resources required for a work, selects an integrated system in which the work is executed from the plurality of integrated systems.