Storlet Workflow Optimization via Virtualization Metadata
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining a compute node to implement computation algorithms in object storage systems are inefficient and do not consider licensing types and guidelines associated with virtualization units, leading to suboptimal resource utilization and increased costs.
Innovation Solution
A method that identifies virtualization unit metadata for each node, including benefits, limitations, and operating costs, to determine an optimal node for implementing computation algorithms based on required operations, thereby optimizing resource allocation and cost management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If current methods are used to determine compute node assignment, then the workflow can be executed, but resource utilization is suboptimal and costs increase
Solution Approach 1:
The system changes the parameters considered in node selection by introducing virtualization unit metadata that includes licensing types, guidelines, benefits, limitations, and operating costs. This transforms the node assignment from a simple availability-based decision to a multi-parameter optimization that considers both performance and cost factors, thereby improving resource utilization efficiency while reducing operating costs.
Solution Approach 2:
The system implements feedback mechanisms by collecting and utilizing virtualization unit metadata about licensing types, guidelines, and operating costs. This feedback information is fed back into the node selection process, enabling continuous optimization of resource allocation that balances performance requirements with cost efficiency, thus improving productivity while reducing energy loss.
2Reliability
If current methods are used for node selection, then computation can be executed, but licensing compliance cannot be ensured
Solution Approach 1:
The system performs preliminary action by pre-collecting and storing virtualization unit metadata that includes licensing types and guidelines before node selection is needed. This advance preparation ensures that licensing compliance can be verified automatically during node assignment, improving reliability without significantly increasing the complexity of the selection process itself.
Solution Approach 2:
The system introduces virtualization unit metadata as an intermediary structure that bridges the gap between licensing requirements and node selection. This intermediary contains all necessary licensing information in a standardized format, enabling automatic compliance verification while keeping the node selection logic relatively simple and maintainable.
3Productivity
If more comprehensive node evaluation criteria are applied, then optimal resource allocation is achieved, but the evaluation process becomes more complex
Solution Approach 1:
The system applies segmentation by dividing the node evaluation into distinct components: virtualization unit metadata (containing licensing info), computation algorithm requirements, and node availability status. This segmentation allows each component to be evaluated independently and then combined through a systematic comparison process, improving resource allocation efficiency while managing complexity through modular evaluation steps.
Solution Approach 2:
The system achieves universality by creating a multi-functional evaluation framework that can handle different computation algorithms, virtualization units, and licensing types through a single unified process. The virtualization unit metadata structure serves multiple purposes: it stores licensing information, provides comparison criteria, and enables automated decision-making, thereby improving resource allocation efficiency without proportionally increasing process complexity.
Data Source
AI summary
A computer-implemented method according to one embodiment includes identifying virtualization unit metadata for each of a plurality of nodes, the virtualization unit metadata including one or more guidelines for an application that is implemented by a virtualization unit, where the one or more guidelines include one or more benefits of a virtualization unit, one or more limitations of the virtualization unit, and one or more operating costs of the virtualization unit; determining, from the plurality of nodes, an optimal node for implementing a computation algorithm based on one or more required computation operations and the virtualization unit metadata for each of the plurality of nodes; and returning an identification of the optimal node.