Container Resource Planning for Replica-Aware Failure Tolerance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In container systems, the manual preparation of worker nodes based on replica set count and affinity rules leads to conflicts that can result in service disruptions when failures occur, as the declarative state cannot be maintained.
Innovation Solution
A container resource design device that automatically calculates the required worker nodes by considering replica set count, affinity rules, and failure tolerance count, dispensing resources to maintain self-healing capabilities and ensure service continuity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the replica set count is increased to maintain service availability, then service continuity is improved, but the number of required worker nodes increases leading to resource waste when failures occur
Solution Approach 1:
The system pre-calculates the minimum required worker node count by considering the replica set count and failure tolerance count before deployment. This preliminary calculation ensures that the exact number of worker nodes needed is allocated, avoiding both over-provisioning and under-provisioning of resources.
Solution Approach 2:
The system dynamically adjusts the worker node allocation based on the parameters of replica set count and failure tolerance count. By changing these parameters, the system can optimize the worker node count to maintain service continuity while minimizing resource waste.
2Productivity
If manual resource design is performed to optimize worker node allocation, then resource efficiency is improved, but the complexity of system configuration and maintenance increases
Solution Approach 1:
The system automatically calculates and allocates the minimum required worker nodes based on the replica set count and failure tolerance count without requiring manual intervention. This self-service approach maintains resource efficiency while eliminating the complexity of manual configuration and maintenance.
Solution Approach 2:
The manual process of calculating and allocating worker nodes is replaced by an automated calculation mechanism. The system uses algorithms to compute the minimum required worker node count, substituting manual mechanical processes with automated computational methods.
3Ease of operation
If affinity rules are strictly enforced to distribute pods across different worker nodes, then load balancing is improved, but the ability to maintain replica set count during failures deteriorates
Solution Approach 1:
The system pre-calculates the minimum required worker node count by considering both the affinity rules for load balancing and the failure tolerance count for reliability. This preliminary calculation ensures that enough worker nodes are allocated to maintain the replica set count even when failures occur, while still enforcing affinity rules for load balancing.
Data Source
AI summary
A master node, which is a container resource design device, includes: a container setting reception unit configured to acquire definition information including a replica set count indicating a number of redundancies of a pod, an affinity rule indicating a condition for starting up containers, and a failure tolerance count; a resource calculation unit configured to calculate a required worker node count while satisfying the replica set count and satisfying the condition defined in the affinity rule, even if a failure of the failure tolerance count or less has occurred in a worker node, and transmit resource dispensation instruction information to a resource management mechanism; and a container setting unit configured to determine placement of the pods on dispensed worker nodes and transmit container setting information.