Container Orchestrator Headroom Simulation for Predictive Scaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current automatic infrastructure scaling for container-based workloads fails to anticipate future demands, leading to resource exhaustion and service degradation due to delays in provisioning additional infrastructure, which results in increased waiting times and reduced quality of service.
Innovation Solution
A method and system that predicts and pre-provisions additional infrastructure by determining properties of headroom containers, using machine learning and user inputs, to ensure availability when needed, thereby avoiding stalls and optimizing resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If current automatic infrastructure scaling monitors and provisions resources immediately upon demand, then infrastructure provisioning speed is improved, but future demand anticipation is lost leading to resource exhaustion
Solution Approach 1:
The system performs preliminary actions by predicting future container demand using machine learning models and proactively provisioning infrastructure resources before they are actually needed. The service controller estimates future container requirements and triggers infrastructure provisioning in advance, ensuring resources are ready when demand occurs, thus avoiding resource exhaustion while maintaining fast provisioning capability.
2Productivity
If infrastructure is provisioned reactively when containers are scheduled, then resource allocation is optimized, but waiting time increases and service quality degrades
Solution Approach 1:
The system estimates future container requirements using machine learning predictions and triggers infrastructure provisioning before containers actually need to be deployed. This preliminary action eliminates the waiting time delay between when containers are scheduled and when infrastructure becomes available, while still optimizing resource allocation by provisioning only what is predicted to be needed.
Solution Approach 2:
The system continuously monitors actual container deployment patterns and compares them against machine learning predictions, using this feedback to refine future predictions. This feedback mechanism improves the accuracy of demand forecasting, allowing the system to provision infrastructure more precisely in advance, reducing both waiting time and resource waste.
3Reliability
If the system provisions additional infrastructure to handle peak demand, then service availability is improved, but infrastructure costs increase
Solution Approach 1:
The system uses machine learning models to predict future container demand with sufficient accuracy to provision infrastructure just in time for peak loads. By anticipating demand patterns rather than over-provisioning for maximum possible demand, the system ensures service availability during peaks while minimizing unnecessary infrastructure resources and associated costs.
Data Source
AI summary
A method, system and computer program product, the method comprising: determining properties of a set of containers that are deployed over a computer infrastructure, wherein the computer infrastructure is provisioned via an infrastructure management service; determining properties of one or more headroom containers, wherein the one or more headroom containers are not deployed over the computer infrastructure; simulating the container orchestrator using the properties of the set of container and the properties of the headroom containers, for obtaining an expected deployment of the set of containers together with the one or more head room containers; based on the expected deployment, determining whether the computer infrastructure is sufficient for deploying the set of containers together with the one or more headroom containers; and subject to the computer infrastructure being insufficient, issuing a request to the infrastructure management service to allocate additional computer infrastructure.