CCN Process Migration via Manifest-Based Resource Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current content centric network (CCN) architectures face challenges in efficiently migrating processes, such as virtual machines, across systems, as they lack effective methods for managing and transferring machine resources and maintaining data consistency during migration.
Innovation Solution
The system employs a pre-copy method with hash-based names and manifests to facilitate process migration by mapping machine elements to CCN names, utilizing interest/content object exchanges and checkpoint manifests to manage resource identification, transfer, and de-duplication, ensuring efficient migration across CCN networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional CCN architectures are used for process migration, then basic content delivery is possible, but efficient migration of virtual machines and management of machine resources cannot be achieved
Solution Approach 1:
The patent segments the process migration into distinct phases (push phase, stop-and-copy phase, pull phase) and breaks down machine resources into individual named components that can be independently managed and transferred through the CCN, enabling efficient migration while maintaining adaptability
Solution Approach 2:
The patent introduces manifests as intermediary data structures that describe machine resources and their names, enabling the CCN to efficiently route and transfer resource data during migration without requiring direct source-destination communication for each resource
2Adaptability or versatility
If process migration is implemented in CCN, then virtual machine migration becomes possible, but data consistency and integrity during migration cannot be ensured
Solution Approach 1:
The patent performs preliminary actions by creating checkpoints of the virtual machine state before migration, naming and describing resources in advance through manifests, and pre-establishing the migration framework so that data consistency can be maintained throughout the migration process
Solution Approach 2:
The patent implements feedback mechanisms where the target system verifies received resource data against the manifest descriptions, ensures data integrity through naming conventions, and provides status information back to the source system to maintain consistency during migration
3Ease of operation
If manual resource mapping is used for migration, then control over resource transfer is possible, but migration complexity and time consumption increase
Solution Approach 1:
The patent enables self-service by allowing the CCN infrastructure to automatically route and transfer resource data based on the names provided in manifests, eliminating the need for manual resource mapping and control while reducing migration time through automated name-based routing
Data Source
AI summary
One embodiment provides a system. In operation, the system generates a first interest for a first manifest which indicates a first phase of a migration of a data model and indicates a first plurality of resources with corresponding names; in response to receiving the first manifest, transmits, based on the corresponding names of the first plurality of resources, a first plurality of interests for the first plurality of resources; receives a second manifest which indicates a second phase of the migration and indicates a second plurality of resources with corresponding names; transmits, based on the corresponding names of the second plurality of resources, a second plurality of interests for the second plurality of resources; and starts the data model on the system based on content objects retrieved in response to the first and second plurality of interests.