Stream Processing Resource Allocation Using Backlog Feedback

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing computer networks face challenges in efficiently allocating resources to process unbounded and unpredictable data streams, leading to issues of overprovisioning (resource waste) and underprovisioning (data backlog and latency), which are not effectively addressed by traditional batch or worst-case scenario planning.

Innovation Solution

A dynamic resource allocation system that uses backlog growth and volume metrics to adjust processing resources iteratively, ensuring that the backlog remains at or below a target level, thereby preventing overprovisioning and underprovisioning by matching resource allocation to the actual workload demands.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional batch or worst-case scenario planning is used to allocate resources, then resource availability is ensured, but resource waste occurs due to overprovisioning

Engineering Contradiction:

Improveresource availabilityVSAvoidresource waste

Core Design Contradiction:

ReliabilityVSLoss of energy

Solution Approach 1:

The patent implements dynamic resource allocation that continuously monitors backlog growth and volume metrics, adjusting the quantity of processing resources allocated to jobs in real-time based on actual workload demands rather than static worst-case scenarios. This allows the system to maintain reliability during high-demand periods while reducing resource waste during low-demand periods.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system employs feedback mechanisms by iteratively determining backlog growth and volume metrics, then using this information to adjust resource allocation. The feedback loop continues by monitoring whether the backlog remains at or below target levels, enabling continuous optimization of resource utilization while ensuring service level agreements are met.

Inventive Principle:

Principle #23Feedback

2Reliability

If traditional batch or worst-case scenario planning is used to allocate resources, then resource availability is ensured, but data backlog and latency increase due to underprovisioning during peak demand

Engineering Contradiction:

Improveservice level agreement complianceVSAvoiddata backlog and latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The dynamic resource allocation system adjusts the quantity of processing resources in real-time based on monitored backlog metrics, enabling the system to provision additional resources during peak demand periods to prevent data backlog and latency, while de-provisioning during low-demand periods to optimize resource utilization.

Inventive Principle:

Principle #15Dynamics

3Productivity

If the quantity of processing resources is increased to handle unbounded data streams, then processing speed improves, but resource waste increases

Engineering Contradiction:

Improveprocessing speedVSAvoidresource waste

Core Design Contradiction:

ProductivityVSLoss of energy

Solution Approach 1:

The system dynamically adjusts the quantity of processing resources allocated to jobs based on real-time backlog growth and volume metrics, increasing resources when processing speed is needed to handle unbounded data streams and reducing resources when backlog is low, thereby optimizing both productivity and resource utilization.

Inventive Principle:

Principle #15Dynamics

4Loss of energy

If the quantity of processing resources is decreased to reduce resource waste, then resource utilization improves, but data backlog and processing latency increase

Engineering Contradiction:

Improveresource utilizationVSAvoiddata backlog and processing latency

Core Design Contradiction:

Loss of energyVSLoss of time

Solution Approach 1:

The dynamic resource allocation system continuously monitors backlog metrics and adjusts resource allocation accordingly, decreasing resources to improve utilization when backlog is low and increasing resources to prevent backlog accumulation and latency when demand increases, achieving optimal balance between resource efficiency and service quality.

Inventive Principle:

Principle #15Dynamics

Data Source

PatentUS12541403B2Resource allocation for computer processing

Publication Date: 2026.02.03 GOOGLE LLC

AI summary

A job that receives as input a data stream is executed. For the job, it is iteratively determined a backlog growth over a first period of time; a backlog volume; and whether to adjust a quantity of processing resources. For each iteration for which the quantity of processing resources allocated to the job are determined to be adjusted, adjusting the quantity of processing resources allocated to the job. For each iteration for which the quantity of processing resources allocated to the job are determined not to be adjusted, maintaining the quantity of processing resources allocated to the job.