Cloud Database Edge Updates for Long-Lived Connection Continuity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Standard KUBERNETES rolling update procedures for cloud databases with long-lived connections result in undesirable downtime and connection loss due to immediate termination of active pods, affecting stateful applications.

Innovation Solution

A modified rolling update strategy with extended grace periods and minimum execution times for containers, allowing existing connections to persist while new pods become ready, followed by controlled termination of old pods.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If standard KUBERNETES rolling update procedures are used to replace pods one by one, then the update process can be automated and simple to implement, but the active connections are immediately terminated causing downtime and connection loss

Engineering Contradiction:
Improveease of update implementationVSAvoidconnection continuity
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The patent applies preliminary action by introducing a grace period configuration before pod termination. The system pre-establishes a time window (grace period) during which active connections should be allowed to complete naturally. This is implemented through the `terminationGracePeriodSeconds` field in the pod specification, which sets up advance conditions for connection handling before the actual termination occurs, preventing immediate connection disruption.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent applies dynamics by making the pod termination behavior adjustable and adaptive. Instead of fixed immediate termination, the system allows dynamic control over the termination timing through configuration parameters. The grace period can be adjusted based on connection characteristics, and the system dynamically adapts its behavior to balance update speed with connection continuity requirements.

Inventive Principle:
Principle #15Dynamics

2Productivity

If standard rolling update procedures terminate pods immediately, then the update speed is fast and productivity is maintained, but long-lived database connections are dropped causing computation time loss

Engineering Contradiction:
Improveupdate speedVSAvoidcomputation time loss
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent applies parameter changes by modifying the termination timing parameter from immediate to configurable grace period. The `terminationGracePeriodSeconds` parameter allows changing the termination behavior from a fixed fast track to an adaptive time-based approach. This parameter adjustment enables the system to maintain update speed while preventing unnecessary computation time loss from premature connection termination.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If extended grace periods are implemented to preserve existing connections, then connection continuity is improved, but the update process takes longer and downtime increases

Engineering Contradiction:
Improveconnection continuityVSAvoidupdate duration
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies partial action by implementing grace periods selectively rather than universally. The system evaluates each pod's connection state and applies the grace period only when necessary (for pods with active long-lived connections), while allowing immediate termination for pods without active connections. This partial application of the grace period mechanism balances connection continuity with update speed.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS20260064402A1Near-zero downtime maintenance of the edge layer in cloud databases
Publication Date: 2026.03.05 SAP SE
  • US20260064402A1 patent drawing
  • US20260064402A1 patent drawing
  • US20260064402A1 patent drawing

AI summary

Near-zero downtime maintenance of containerized applications can be achieved via a modified rolling update strategy for a container orchestration platform. During deployment of a first set of containers based on a first deployment that specifies an image of a first software version and a grace period for preserving existing connections to the first set of containers, a second deployment object is received that specifies an image of a second software version and a minimum execution time for a second set of containers after which new connections to the second set of containers are allowed. After deployment of the second set of containers, new connections to the second set of containers are not enabled until the respective containers have executed for the minimum execution time. Existing connections to each container of the first set of containers are preserved until the earlier of their completion or expiration of the grace period.