Microservice Scheduling via Network State Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current container scheduling approaches in software-defined data centers are sub-optimal as they fail to adequately address the dynamic nature of applications, leading to unwanted latencies and performance issues due to the lack of consideration for network states and resource utilization across virtual computing instances and hypervisors.
Innovation Solution
A method for microservice scheduling that takes into account network states, including VCI network states, hypervisor network statistics, and application latencies, to deploy containers and microservices efficiently across virtual computing instances and hypervisors, using a microservice scheduling agent that collects and analyzes network state information to optimize resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If container scheduling is performed using traditional approaches that do not consider network states, then the scheduling process is simple and fast, but application performance deteriorates due to unwanted latencies and resource inefficiencies
Solution Approach 1:
The system performs preliminary collection and analysis of network state information (hypervisor network statistics, VCI network states) before scheduling containers. This advance preparation enables the scheduler to make informed decisions about optimal container placement that consider network conditions, thereby reducing application latency and improving execution speed without sacrificing scheduling simplicity
2Reliability
If network state information is collected and analyzed for scheduling decisions, then application performance and resource allocation are optimized, but system complexity increases due to additional data collection and processing requirements
Solution Approach 1:
The scheduling system integrates multiple functions into a unified approach: network state collection, network state analysis, and container scheduling all occur within a single coordinated process. The scheduler simultaneously considers resource availability and network conditions, eliminating the need for separate complex subsystems and reducing overall system complexity while maintaining reliable resource allocation
Solution Approach 2:
The system continuously monitors network states and uses this feedback information to optimize scheduling decisions. By analyzing hypervisor network statistics and VCI network states in real-time, the scheduler adjusts container placement dynamically, improving resource allocation efficiency without requiring overly complex predetermined rules
Data Source
AI summary
A method for containerized workload scheduling can include determining a network state for a first hypervisor in a virtual computing cluster (VCC). The method can further include determining a network state for a second hypervisor. Containerized workload scheduling can further include deploying a container to run a containerized workload on a virtual computing instance (VCI) deployed on the first hypervisor or the second hypervisor based, at least in part, on the determined network state for the first hypervisor and the second hypervisor.