Cloud Management System Edge Node Segmentation for Latency Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud management systems (CMS) face significant challenges in managing the high monitored-data load and processing load/delay, which can lead to inefficient resource allocation and quality of service (QoS) issues due to the sheer volume of data and complexity in Network Function Virtualization (NFV) environments.
Innovation Solution
The implementation of a prediction process to provide predicted key performance indicator (KPI) values for VMs, using machine learning techniques to assess accuracy and adjust monitoring frequency, and proactive placement of VMs to reduce data exchange and processing loads, thereby optimizing resource utilization and decision-making processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a CMS monitors and processes all monitored data from multiple datacenters to make LCM decisions, then the quality and accuracy of LCM decisions is improved, but the processing load and delay on the CMS increases significantly
Solution Approach 1:
The patent segments the monolithic CMS into distributed edge computing nodes deployed at each datacenter. These edge nodes perform local monitoring, data processing, and preliminary LCM decisions, reducing the volume of data and decision-making burden on the central CMS. This segmentation enables parallel processing across multiple nodes, thereby reducing overall processing delay while maintaining decision accuracy through coordinated operation.
Solution Approach 2:
The patent implements preliminary filtering and aggregation of monitored data at edge computing nodes before transmitting to the central CMS. Edge nodes perform initial analysis, anomaly detection, and data preprocessing, so that only critical or aggregated information reaches the central system. This preliminary action reduces the processing load on the central CMS and accelerates response time for time-sensitive LCM decisions.
2Reliability
If the CMS collects and analyzes large volumes of monitored data from all datacenters, then the reliability of resource utilization information is improved, but the monitoring load and network bandwidth consumption increase
Solution Approach 1:
The patent enables each edge computing node to perform local data filtering and quality assessment, adapting the monitoring granularity and data collection frequency to local conditions and requirements. This local quality approach ensures that each datacenter contributes only the most relevant and reliable data to the central CMS, reducing overall data volume while maintaining the reliability needed for accurate LCM decisions.
Solution Approach 2:
Edge computing nodes perform preliminary data validation, aggregation, and filtering before transmitting information to the central CMS. This preliminary processing ensures data quality and reliability at the source, eliminating the need for the central system to process redundant or low-quality data, thereby reducing network bandwidth consumption and central processing load.
3Speed
If the CMS performs frequent monitoring and analysis of VNF KPIs and resource utilization, then the responsiveness to LCM events is improved, but the processing load and energy consumption increase
Solution Approach 1:
The patent implements dynamic monitoring frequency adjustment at edge computing nodes based on local system conditions, event severity, and historical patterns. During normal operation, monitoring frequency is reduced to conserve energy; when anomalies or critical events are detected, the system automatically increases monitoring intensity. This dynamic approach maintains rapid response capability for critical events while significantly reducing average processing load and energy consumption.
Solution Approach 2:
The patent employs periodic monitoring with variable intervals at edge nodes, where the monitoring period adapts based on system state. Instead of continuous high-frequency monitoring, the system uses event-triggered periodic sampling that intensifies during critical periods and relaxes during stable operation. This periodic action with adaptive intervals maintains response speed for important events while reducing overall processing energy requirements.
Data Source
AI summary
According to an embodiment of the invention, a method is provided for reducing the monitored-data load and the processing load/delay on a cloud management system (CMS). The method includes applying a prediction process to provide predicted key performance indicator (KPI) values for a plurality of VMs managed by the CMS during a first monitoring epoch; collecting, during the first monitoring epoch, observed KPI values for the plurality of VMs managed by the CMS; assessing the accuracy of the prediction process by way of calculating, according to a reward function, reward values for the plurality of VMs based on a deviation between the observed KPI values and the predicted KPI values; calculating a monitoring frequency for collecting monitoring information during a second monitoring epoch based on the reward values; and collecting the monitoring information during the second monitoring epoch according to the calculated monitoring frequency.