Dynamic Resource Allocation for Asynchronous Data Workflows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cloud provider networks face challenges in dynamically managing data traffic workflows to meet performance requirements, such as service level agreements, due to dependencies between operations and changing workloads, leading to potential underutilization or overutilization of resources.
Innovation Solution
An asynchronous data traffic workflow management system coordinates the performance of data traffic workflows across multiple systems, utilizing target capacity generation to dynamically adjust resource allocation based on pending tasks and performance requirements, ensuring that resources are rightsized to meet performance demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If static resource allocation is used for data traffic workflows, then system simplicity is maintained, but performance requirements cannot be dynamically met and resources are either underutilized or overutilized
Solution Approach 1:
The patent implements dynamic performance configuration by allowing the workflow management system to adjust resource allocation in real-time based on pending tasks and performance requirements. The system transitions from static to dynamic resource management, enabling automatic scaling of computing devices or service instances according to workload demands, thereby resolving the contradiction between resource utilization efficiency and system complexity.
Solution Approach 2:
The system incorporates feedback mechanisms where performance data from workflow execution is continuously monitored and fed back to the workflow management system. This feedback loop enables the system to automatically adjust resource allocation based on actual performance metrics, meeting performance requirements while optimizing resource utilization without manual intervention.
2Loss of time
If resource allocation is increased to meet performance requirements, then task completion time is reduced, but resource waste occurs when performance demands are not met
Solution Approach 1:
The patent dynamically changes resource allocation parameters based on actual workflow performance requirements. The system adjusts the number of computing devices or service instances allocated to a workflow based on the number of pending tasks and performance targets, ensuring resources are increased only when needed to meet performance requirements and reduced when demands are lower, thereby avoiding resource waste.
Solution Approach 2:
The system implements dynamic resource scaling that responds to real-time workflow characteristics. By continuously monitoring pending tasks and performance requirements, the system dynamically adjusts resource allocation to match actual demands, preventing both over-provisioning (resource waste) and under-provisioning (task completion delays).
3Adaptability or versatility
If multiple services are coordinated to implement features, then functionality is enhanced, but performance coordination becomes complex and difficult to manage
Solution Approach 1:
The patent introduces a workflow management system as an intermediary layer that coordinates performance across multiple services. This intermediary abstracts the complexity of inter-service performance coordination by providing a unified interface for managing workflows, allocating resources, and monitoring performance across service boundaries, thereby enabling enhanced functionality without proportionally increasing coordination complexity.
Solution Approach 2:
The workflow management system implements universal functionality that can handle various workflow types and service combinations through a single coordinated framework. By providing multi-functional capabilities for task management, resource allocation, and performance monitoring across different services, the system reduces the need for separate coordination mechanisms for each service combination.
Data Source
AI summary
The performance of asynchronous tasks may be dynamically configured. An evaluation of pending tasks that includes updates for new tasks and completed tasks being removed is performed. A target capacity for the data traffic workflow is determined from the evaluation of pending tasks in order to satisfy a performance requirement for the data traffic workflow. Modifications to a performance configuration for the data traffic workflow are then made based on a comparison with the target capacity.