Software-Defined Data Center Drift Detection via Proactive and Passive Polling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In software-defined data centers (SDDCs), detecting drift from the desired state is computationally expensive due to the need for frequent compliance checks, which can lead to delays in identifying configuration changes.
Innovation Solution
Implementing a method for automatic drift detection in SDDC configurations using proactive and passive detection mechanisms, where changes in configurations trigger immediate or periodic drift computations, respectively, and notifying a desired state management service of the computed drift.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequent compliance checks are performed to detect drift from desired state, then measurement precision is improved, but use of energy is worsened due to computational expense
Solution Approach 1:
The system dynamically adjusts drift detection strategies based on change notifications. When services are known to be stable, passive periodic checks are used. When change notifications are received, proactive drift computation is triggered. This dynamic adaptation optimizes computational resources while maintaining drift detection accuracy.
Solution Approach 2:
The system implements feedback mechanisms where drift computation results and change notifications inform future detection behavior. The drift poller uses feedback from service change notifications to adjust polling frequency and trigger proactive drift computations, creating a closed-loop system that reduces unnecessary computational expense while maintaining detection precision.
2Loss of energy
If compliance check frequency is reduced to lower computational costs, then use of energy is improved, but loss of time is worsened due to delayed drift detection
Solution Approach 1:
The system performs preliminary drift computations proactively when change notifications are received, before actual drift occurs or is detected by periodic polling. This preliminary action ensures that drift is detected immediately upon configuration changes without waiting for the next scheduled compliance check, reducing detection delay while maintaining lower overall computational frequency.
Solution Approach 2:
The system implements periodic passive drift polling at extended intervals between proactive computations. This periodic action maintains a baseline detection capability while significantly reducing computational frequency compared to continuous or frequent polling, thereby lowering energy consumption while preventing excessive detection delays through the periodic checks.
3Reliability
If proactive drift detection is implemented for services with change notifications, then reliability is improved through timely drift detection, but device complexity is worsened due to additional detection mechanisms
Solution Approach 1:
The system segments services into different drift detection categories: services with change notification capabilities that receive proactive drift computation, and services without such capabilities that receive passive periodic polling. This segmentation allows the system to apply complex proactive detection only where necessary, improving reliability for critical services while limiting overall system complexity through differentiated treatment of service groups.
Solution Approach 2:
The drift poller acts as an intermediary component that coordinates between service change notifications and drift computation operations. It receives notifications from services, manages the queue of drift computations, and triggers proactive or passive detection appropriately. This intermediary simplifies the overall architecture by centralizing the detection coordination logic rather than requiring direct complex interactions between all services and the drift detection system.
Data Source
AI summary
Drift is automatically detected in configuration of services running in a management appliance of a software-defined data center. A method of automatically detecting drift includes: in response to a notification of a change in a configuration of a first service enabled for proactive drift detection, transmitting a first request to compute drift in the configuration of the first service to a plug-in of the first service, the first request including the change in the configuration of the first service; periodically, at designated time intervals, transmitting a second request to compute drift in the configuration of a second service enabled for passive drift detection, to the plug-in of the second service, the second request including a current state of the configuration of the second service; and notifying a desired state management service of the computed drift in the configuration of the first and second services.