Proactive Autoscaling via Predictive ML Metrics

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional autoscaling methods are reactive and limited to specific platforms and use cases, failing to proactively adjust computing resources in response to anticipated changes in load levels.

Innovation Solution

A proactive autoscaling system utilizing heuristics and machine learning to predict scaling events by gathering real-time metrics from various sources, both internal and external to applications, and dynamically adjusting computing resources before actual scaling events occur, regardless of the platform, use case, or type of resources.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional reactive autoscaling is used, then resources are allocated in response to detected load changes, but the system cannot proactively adjust resources before scaling events occur

Engineering Contradiction:

Improvesystem performanceVSAvoidresponse time to load changes

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by proactively scaling computing resources before scaling events actually occur. The proactive auto-scaler predicts future scaling events using machine learning and heuristics, and executes resource allocation in advance, rather than waiting to detect and respond to load changes after they occur.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by tracking empirical data over time, including gleaned metrics, predictions of scaling events, occurrences of predicted scaling events, and results of scaling actions. This feedback loop is used to refine heuristics and improve machine learning models, enabling more accurate predictions and better resource allocation decisions.

Inventive Principle:

Principle #23Feedback

2Reliability

If conventional platform-specific auto-scalers are used, then autoscaling is optimized for specific use cases and platforms, but the system lacks universality across different platforms and protocols

Engineering Contradiction:

Improveautoscaling accuracyVSAvoidplatform independence

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The proactive auto-scaler is designed as a universal system that can operate across multiple platforms, ecosystems, and protocols. It gleans metrics from diverse sources including applications, electronic news media, social media, email, and text messages, and can predict and respond to scaling events regardless of the specific platform or communication protocol being used.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system acts as an intermediary layer between diverse data sources and the resource allocation mechanisms. The proactive auto-scaler collects and processes metrics from various platforms and protocols, translates them into predictions, and then executes resource scaling actions, thereby bridging the gap between platform-specific data and universal resource management.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If multiple conventional auto-scalers for different use cases are deployed, then each use case is addressed, but the overall system complexity increases

Engineering Contradiction:

Improveuse case coverageVSAvoidsystem architecture

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system merges multiple use-case-specific auto-scaling functions into a single unified proactive auto-scaler. By combining metric collection from diverse sources, predictive analytics using machine learning and heuristics, and resource allocation capabilities into one system, it eliminates the need for multiple separate auto-scalers while maintaining versatility across different use cases.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The unified proactive auto-scaler performs multiple functions: collecting metrics from various sources, predicting scaling events using machine learning and heuristics, and executing resource allocation across different platforms and protocols. This multi-functional design reduces system complexity while maintaining broad adaptability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11226844B1Universal, proactive, and dynamic scaling of computing resources

Publication Date: 2022.01.18 DIALPAD INC

AI summary

A proactive autoscaling system can use heuristics and machine learning to proactively, dynamically and automatically scale computing resources allocated to applications up and down, prior to scaling events that cause changes in load levels. The proactive autoscaling system may be stateless, and may be agnostic to use case, platform, field of endeavor, or communication protocol used by the applications. The proactive autoscaling system gleans metrics in real-time. The gleaned metrics are indicative of load levels concerning one or more applications. These gleaned metrics may be in a variety of formats, and may be from different sources, both internal or external to the applications. The proactive autoscaling system automatically predicts scaling events based on gleaned metrics. Prior to the occurrence of a predicted scaling event, the proactive autoscaling system can automatically scale computing resources available to one or more target applications, in response to the predicting of the scaling event.