Network Slice Management via Centralized Datastore
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual network slicing technologies require duplicating network configuration data for each virtual network slice, leading to overhead and inefficiencies in managing and configuring multiple logical networks on a shared physical infrastructure.
Innovation Solution
Implementing a centralized network management entity that maintains a single physical network datastore, using classification and run-time allocation methods to generate slice views and configure virtual network slices without duplicating data, through the use of XSLT transformations and metadata for filtering and tagging network entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If network configuration data is duplicated for each virtual network slice, then each slice can be independently configured and managed, but memory usage and data overhead increase significantly
Solution Approach 1:
The patent merges the configuration data of multiple virtual network slices into a single shared physical network datastore. Instead of maintaining separate configuration copies for each slice, the system consolidates all configuration data in one location, reducing memory usage while enabling efficient slice-specific views through selective data retrieval and presentation.
Solution Approach 2:
The shared physical network datastore serves multiple functions: it stores configuration data for the entire physical network infrastructure and simultaneously supports multiple virtual network slices. This universal datastore structure allows the same data storage mechanism to serve diverse slice configuration needs without duplication, achieving multi-functionality with a single system component.
2Ease of operation
If network configuration data is duplicated for each virtual network slice, then slice-specific management is simplified, but data synchronization and consistency become more complex
Solution Approach 1:
By merging all network configuration data into a single shared datastore, the patent eliminates the complexity of synchronizing data across multiple duplicate configurations. The unified structure ensures data consistency automatically, as there is only one source of truth for the physical network configuration that all virtual slices reference.
Solution Approach 2:
The patent introduces an intermediary mechanism that sits between the shared physical network datastore and individual virtual network slices. This intermediary layer handles slice-specific data retrieval, filtering, and presentation, simplifying slice management while maintaining data consistency. The intermediary acts as a mediator that translates unified datastore content into slice-specific views without requiring data duplication.
3Reliability
If a centralized network management entity maintains a single physical network datastore, then data integrity is improved, but access control and slice isolation become more difficult
Solution Approach 1:
The patent employs an intermediary access control mechanism that manages retrieval and presentation of data from the centralized datastore to different virtual network slices. This intermediary layer enforces slice isolation by filtering and selectively presenting only the configuration data relevant to each slice, while maintaining the integrity of the centralized datastore. The intermediary acts as a gatekeeper that preserves data integrity while enabling controlled access.
Solution Approach 2:
The system applies local quality control by providing each virtual network slice with a customized view of the centralized datastore. While the underlying datastore remains unified and intact, each slice receives tailored access to specific configuration subsets appropriate to its needs. This local quality approach maintains centralized data integrity while enabling slice-specific access control and isolation.
4Quantity of substance
If network configuration data is not duplicated across slices, then memory efficiency improves, but run-time allocation and assignment of entities to slices becomes necessary
Solution Approach 1:
The patent applies preliminary action by pre-organizing and tagging configuration data in the shared physical network datastore with metadata that identifies slice associations. This preliminary structuring of data enables efficient run-time allocation and assignment of network entities to slices without requiring data duplication. The pre-established organizational framework allows rapid retrieval and assignment operations.
Solution Approach 2:
The patent introduces an intermediary allocation mechanism that efficiently manages the assignment of network entities to virtual slices at run-time. This intermediary layer handles the dynamic allocation process by querying the shared datastore and assigning appropriate configuration data to slices based on current network state and requirements, maintaining both memory efficiency and configuration productivity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods, apparatuses and computer readable storage mediums provide virtual network slicing without duplicating network configuration data (also referred to as network configuration information) on a slice-by-slice basis by maintaining a single physical network datastore including network configuration information for all network entities in the physical network, but generating slice views for respective virtual network slices as needed over time. Methods, apparatuses and computer readable storage mediums also enable configuration of a plurality of virtual network slices sharing a physical network infrastructure.