Cross-Workgroup Node Allocation for Predicted Idle Transaction Windows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network resources within computing networks often face fluctuating workloads that exceed their capacity, leading to increased costs when additional resources are purchased from service providers, and existing methods do not efficiently manage workload distribution across different work groups.
Innovation Solution
A node selection system uses machine learning to predict idleness durations and runtimes, allowing nodes from one work group to borrow resources from another, optimizing parallel execution and managing work group credits to handle interruptions and workload spikes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If network resources are increased to address workload spikes, then the capacity to handle transactions is improved, but the cost to the company increases
Solution Approach 1:
The patent enables network resources to serve multiple work groups dynamically. A node dedicated to one work group can execute transactions from another work group when idle, making the resource universal and multi-functional. This resolves the contradiction by allowing existing resources to handle workload spikes from any work group without requiring dedicated resources for each, thus maintaining productivity while avoiding increased costs.
Solution Approach 2:
The system dynamically changes the parameter of resource allocation by predicting idle periods and assigning transactions during those periods. Instead of static resource allocation, the system adjusts resource usage parameters based on predicted workload patterns, enabling existing resources to handle peak loads without permanent capacity increases, thereby maintaining productivity without increasing cost.
2Reliability
If network resources are dedicated to specific work groups, then transaction execution reliability is improved, but resource utilization efficiency deteriorates
Solution Approach 1:
The patent introduces dynamic resource allocation where nodes can switch between dedicated and shared modes. During predicted idle periods, nodes dynamically accept transactions from other work groups, while maintaining their primary dedication. This dynamic approach ensures reliability for primary work groups while improving overall resource utilization efficiency by preventing idle resources from going unused.
Solution Approach 2:
The system uses machine learning models that run on the network resources themselves to predict idle periods and automatically allocate transactions. The resources essentially serve themselves by identifying and utilizing their own idle capacity for additional work, improving utilization efficiency without compromising the reliability of their primary function.
3Productivity
If additional network resources are purchased from service providers, then the capacity to handle fluctuating workloads is improved, but the cost increases
Solution Approach 1:
The patent merges the resource pools of multiple work groups into a shared infrastructure. Instead of each work group having separate dedicated resources, the system combines resources and uses machine learning to intelligently allocate them based on predicted idle periods. This merging allows the organization to handle fluctuating workloads across all work groups using a single resource pool, eliminating the need to purchase additional resources for each work group and thereby reducing costs while maintaining productivity.
Data Source
AI summary
A computing network includes nodes of different work groups. Nodes of a work group are dedicated to transactions of the work group. If a node of a first work group is predicted to have an idleness window, a second work group may borrow the node to execute a transaction of the second work group. At least a subset of steps of the transaction may be categorized into a step group. Trees of a transaction may be categorized into one or more tree groups. A node is selected for executing a transaction, if the predicted idleness duration of the node is sufficient relative to the predicted runtime of the transaction, the step group, and/or tree group. A credit system is maintained. A first work group transfers a credit to a second work group when borrowing a node of the second work group for executing a transaction of the first work group.