Resource Provisioning Estimator for Service-Oriented Architectures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional resource provisioning techniques for Service-Oriented Architectures (SOAs) are inadequate due to the complexity and granularity of services, leading to suboptimal results, as they fail to accurately account for the increased knowledge of service connections and workflows, resulting in potential bottlenecks and inefficient hardware utilization.
Innovation Solution
A system and method for estimating resource provisioning in SOAs, which includes a model manager, workflow manager, propagated workload engine, and provisioning engine that traverse workflow paths to determine service node workloads and propagate workloads across service nodes, ultimately provisioning services onto hardware nodes based on attribute values and multipliers, ensuring accurate resource allocation and utilization analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional resource provisioning techniques are used for SOAs, then the provisioning process is simple, but the estimation accuracy is insufficient due to failure to account for service connections and workflows
Solution Approach 1:
The patent segments the SOA system into distinct components: service nodes, hardware nodes, workflow paths, and node connectors. This segmentation allows the provisioning estimator to analyze each component's resource requirements separately while maintaining their interrelationships, thereby improving estimation accuracy without overwhelming complexity
Solution Approach 2:
The patent introduces a provisioning estimator as an intermediary system that mediates between the complex SOA architecture and resource provisioning decisions. This intermediary captures service connections and workflows through structured models (service node graphs and workflow paths) to enable accurate resource estimation without requiring direct analysis of the entire complex system
2Adaptability or versatility
If service granularity is increased in SOA, then system flexibility is improved, but resource provisioning complexity increases
Solution Approach 1:
The patent creates a universal provisioning estimator that can handle diverse service types and configurations through a unified modeling approach. The service node graph and workflow path structures provide a multi-functional framework that adapts to different service granularities and SOA configurations, maintaining flexibility while standardizing the provisioning process
Solution Approach 2:
The patent uses parameter-based modeling where services are characterized by adjustable parameters (resource requirements, connection types, workflow roles). This allows the provisioning system to adapt to different service granularities by modifying parameters rather than restructuring the entire provisioning approach, thus maintaining flexibility without proportional increases in complexity
3Reliability
If service connections and workflows are fully modeled, then bottleneck detection is improved, but the provisioning process becomes more complex
Solution Approach 1:
The patent performs preliminary modeling of service connections and workflows before resource provisioning. By establishing service node graphs and workflow paths in advance, the system identifies potential bottlenecks and resource constraints before deployment, improving reliability without adding complexity during the actual provisioning execution phase
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method and system are described for estimating resource provisioning. An example method may include obtaining a workflow path including an external invocation node and respective groups of service nodes, node connectors, and hardware nodes, and including a directed ordered path indicating ordering of a flow of execution of services associated with the service nodes, from the external invocation node, to a hardware node, determining an indicator of a service node workload based on attribute values associated with a service node and an indicator of a propagated workload based on combining attribute values associated with the external invocation node and other service nodes or node connectors preceding the service node in the workflow path based on the ordering, and provisioning the service node onto a hardware node based on combining the indicator of the service node workload and an indicator of a current resource demand associated with the hardware node.