Method for deploying at least one workflow on a computational network of computing systems and deployment control system

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing serverless workflows face significant communication overhead and latency issues due to intermediary communication between function instances, and provisioning new instances incurs a time penalty, which can be inefficient compared to dedicated infrastructure solutions.

Innovation Solution

A method for deploying workflows that dynamically chooses between serverless and monolithic paradigms based on performance parameters, allowing hybrid deployment scenarios to optimize latency and resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If serverless workflow is deployed with multiple function instances, then resource efficiency and scalability are improved, but communication overhead and latency increase

Engineering Contradiction:

Improveresource efficiencyVSAvoidcommunication overhead and latency

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent merges multiple serverless function instances into a single monolithic workflow execution environment. Instead of deploying separate function instances that communicate through intermediaries, the workflow is consolidated into one execution unit that processes all functions sequentially, eliminating inter-service communication overhead and reducing latency while maintaining resource efficiency.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent segments the deployment decision into two distinct modes (serverless vs. monolithic) based on performance parameters. The system evaluates metrics such as execution time, resource usage, and communication overhead to determine the optimal deployment mode, allowing dynamic selection between distributed and consolidated approaches.

Inventive Principle:

Principle #1Segmentation

2Adaptability or versatility

If serverless function instances are provisioned dynamically, then adaptability to workload changes is improved, but provisioning time penalty increases latency

Engineering Contradiction:

Improveadaptability to workload changesVSAvoidprovisioning time penalty

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent performs preliminary provisioning of a single monolithic workflow execution environment before workload changes occur. Instead of provisioning individual function instances on-demand, the system pre-provisions a comprehensive execution environment that can handle varying workloads without requiring additional provisioning time, thus eliminating the latency penalty associated with dynamic instance creation.

Inventive Principle:

Principle #10Preliminary action

3Productivity

If monolithic workflow is deployed on dedicated infrastructure, then performance and resource efficiency are improved, but deployment flexibility and scalability are reduced

Engineering Contradiction:

ImproveperformanceVSAvoiddeployment flexibility

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent introduces dynamic deployment mode selection that allows the system to adapt between serverless and monolithic paradigms based on performance parameters. The deployment configuration is made dynamic, enabling automatic switching between flexible serverless deployment (for high adaptability needs) and optimized monolithic deployment (for performance-critical workloads), thus reconciling flexibility and performance requirements.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentEP4600820A1Method for deploying at least one workflow on a computational network of computing systems and deployment control system

Publication Date: 2025.08.13 SIEMENS AG

Figure 1~2

AI summary

The method for deploying at least one workflow on a computational network comprises the steps of monitoring a performance parameter of the computational network and deploying the workflow in the form of a serverless workflow or in the form of a monolithic workflow, wherein the form of the workflow is chosen depending on the performance parameter.