Subsystem Design Pattern Matching for Large Scale Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current tools and methods in enterprise IT environments are inadequate for designing and managing large-scale systems, making it an art rather than a science, and only a few skilled designers can build highly available systems, which are not mainstream due to scarcity and high costs.
Innovation Solution
The approach involves breaking down large-scale systems into subsystems based on specific characteristics like availability, reliability, and performance, matching each subsystem with a suitable design pattern, and using a management tool to perform tasks such as deployment, monitoring, and resource allocation, facilitated by a meta-pattern analysis that identifies useful design patterns and frameworks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current enterprise IT tools are used for large-scale systems, then general availability is maintained, but system reliability and scalability deteriorate
Solution Approach 1:
The patent divides a large-scale system into multiple subsystems, each with specific characteristics (availability, reliability, statefulness, performance). Each subsystem is matched with appropriate design patterns and managed independently, transforming an unmanageable monolithic system into manageable modular components.
Solution Approach 2:
The patent applies different design patterns and management approaches to different subsystems based on their specific characteristics. For example, stateless subsystems may use different patterns than stateful subsystems, allowing each part of the system to be optimized for its specific requirements rather than applying a one-size-fits-all approach.
2Reliability
If specialized skilled designers are used to build highly available systems, then system availability improves, but cost increases due to scarcity of expertise
Solution Approach 1:
The patent changes the parameters of system design by introducing characteristic-based classification and design pattern matching. This transforms the design process from requiring rare specialized skills to using systematic, repeatable methodologies that can be applied by mainstream development teams.
Solution Approach 2:
The patent creates reusable design patterns that can be copied and applied across multiple subsystems. These patterns encapsulate best practices for building highly available systems, allowing organizations to leverage proven solutions without requiring deep specialized expertise for each new system.
3Productivity
If subsystems are separated based on characteristics, then system scalability improves, but management complexity increases
Solution Approach 1:
The patent introduces a universal management tool that can handle multiple types of subsystems with different characteristics. This tool provides standardized interfaces and workflows for deployment, configuration, monitoring, and repair across diverse subsystems, reducing management complexity despite increased system scalability.
4Productivity
If design patterns are matched to each subsystem, then system performance improves, but initial design time increases
Solution Approach 1:
The patent performs preliminary classification of subsystems based on their characteristics and pre-matches them with appropriate design patterns. This upfront work, potentially automated by the management tool, reduces design time in the long run by avoiding trial-and-error approaches and ensuring optimal patterns are selected from the beginning.
Data Source
AI summary
Described is a technology by which a system corresponding to a large scale application is built from subsystems that are differentiated from one another based on characteristics of each subsystem. Example characteristics include availability, reliability, redundancy, statefulness and/or performance. Subsystems are matched to known design patterns, based on each subsystem's individual characteristics. Each subsystem's characteristics are associated with that subsystem for subsequent use in operation of the system, e.g., for managing/servicing the subsystem. The known design patterns may be provided in a library, in a programming framework, in conjunction with a development tool, and/or as data associated with one or more operating system services, server systems and/or hosted services that include at least one configuration, policy and or schema. Certain design patterns and/or characteristics patterns may be blocked to prevent their usage.