Workflow Activation Policies for Latency Throughput Trade-offs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing workflow tools face challenges in managing large volumes of items through workflows efficiently, as singleton tools process items individually leading to high latency and resource inefficiency, while bulk tools struggle with latency issues when handling multiple items simultaneously.
Innovation Solution
Implementing a 'catch and release' mechanism that allows activities to process items in subsets, enabling efficient chunking and processing of input sets throughout the workflow, thereby reducing latency and improving throughput.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If singleton workflow tools process items individually, then each item receives personalized processing, but throughput is low and latency is high
Solution Approach 1:
The patent divides the input set into multiple subsets and processes them through different paths in the workflow. Some subsets are processed individually through singleton-style activities while others are processed in bulk through bulk-style activities, allowing the system to achieve both personalized processing and high throughput simultaneously
Solution Approach 2:
The workflow system dynamically selects between singleton and bulk processing modes based on the specific activity type and subset characteristics. The system can adaptively switch between processing modes to optimize for either latency or throughput depending on the current workflow state and requirements
2Productivity
If bulk workflow tools process multiple items simultaneously, then throughput is high and resource efficiency improves, but latency increases for individual items
Solution Approach 1:
The patent segments the bulk input set into smaller subsets that can be processed through different workflow paths. This allows the system to maintain high throughput by processing multiple subsets in parallel while reducing individual item latency by routing time-sensitive items through optimized paths
Solution Approach 2:
Different subsets of items receive different processing qualities based on their specific requirements. Time-sensitive items receive prioritized processing with lower latency, while other items are processed in standard bulk mode, allowing the system to optimize local processing characteristics for different item groups
3Quantity of substance
If workflow tools manage large volumes of items, then processing capacity increases, but resource utilization becomes inefficient
Solution Approach 1:
The patent segments large volumes of items into manageable subsets that can be processed efficiently through appropriate workflow paths. This prevents resource exhaustion while maintaining high processing capacity by distributing the load across multiple parallel processing streams
Solution Approach 2:
The system processes items in partial batches rather than attempting to process all items simultaneously. This allows the system to maintain high overall processing capacity while using resources efficiently by completing some processing cycles before starting others, preventing resource overload
Data Source
AI summary
Activation policies for workflows are described. Some implementations support receiving a definition of a workflow, including data defining an input set for the workflow and data defining activities and a flow of the activities for the workflow. The data includes an activation policy for at least a particular activity of the activities. An estimated operational outcome for the workflow and/or each of the activities is determined and displayed. Then data identifying edits to the activation policy are received, and an updated version of the estimated operational outcome for the workflow and/or each of the activities is determined and displayed.