Live Container Migration via Page Table Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current virtualization systems face significant service disruptions and prolonged migration times during hypervisor updates or container migrations, which can negatively impact application performance due to the need for system reboots, lengthy migration processes, and increased network traffic.
Innovation Solution
The method involves identifying containers or processes on virtual machines, determining communication paths, and migrating them to new virtual machines with updated hypervisors or operating systems using advanced page table and warp table mappings to minimize downtime and network traffic, allowing for live updates and replacements without disrupting user interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If live migration is performed to avoid system reboot, then service disruption is reduced, but migration time increases significantly
Solution Approach 1:
The patent segments the migration process into two distinct phases: (1) a fast path for migrating only the necessary container state data, and (2) a slow path for copying remaining memory pages. This segmentation allows the system to quickly migrate critical container information while handling the bulk of data transfer in the background, thereby reducing perceived service disruption without significantly increasing total migration time.
Solution Approach 2:
The patent performs preliminary actions by pre-establishing communication paths between containers and hypervisors before migration occurs. The system determines communication paths in advance and prepares the destination environment, allowing the actual migration to proceed rapidly without unexpected delays. This preliminary preparation enables the migration to complete in under 10 milliseconds for critical paths.
2Duration of action of stationary object
If iterative memory copying is used for live migration, then migration downtime is reduced, but total migration time increases
Solution Approach 1:
The patent determines communication paths between containers and hypervisors in advance before the migration actually occurs. By pre-establishing these paths and preparing the destination environment, the system eliminates the need for iterative memory copying during the critical migration window. The actual migration then proceeds as a single fast operation rather than through multiple iterative copies, reducing both downtime and total time.
Solution Approach 2:
The patent uses a targeted copying approach where only the essential container state data is copied during the fast path migration, rather than copying all memory pages iteratively. This selective copying of only necessary information enables rapid migration while the remaining data is handled through the pre-established communication paths, significantly reducing total migration time while maintaining low downtime.
3Ease of manufacture
If cold container migration is used, then migration is simple, but service disruption occurs
Solution Approach 1:
The patent maintains continuity of useful action by keeping the container running during the migration process. Instead of suspending the container as in cold migration, the system migrates the container state while it continues to execute its workload. The communication paths are established and data is transferred in the background, allowing the container to remain active and serve requests throughout the migration, thereby eliminating service disruption while maintaining migration simplicity.
4Speed
If precopy-based live migration is used, then migration speed increases, but network traffic increases
Solution Approach 1:
The patent extracts only the essential container state data for migration rather than transferring all memory pages. By identifying and migrating only the necessary information (such as process states, memory mappings, and I/O positions), the system achieves fast migration speed while minimizing the volume of data transmitted over the network. This extraction of critical information reduces network traffic significantly compared to transferring complete memory images.
Solution Approach 2:
The patent introduces communication paths as intermediaries between the source and destination hypervisors. These pre-established communication paths enable efficient, targeted data transfer of only necessary container state information, rather than performing broad network transfers. The intermediaries facilitate selective data exchange, allowing fast migration of essential data while avoiding unnecessary network traffic that would occur with complete memory copying.
Data Source
AI summary
Systems and methods for live updating virtualization systems are disclosed. One method may include identifying a container on a first virtual machine of a virtualization system. The container may include at least one process associated with performing a task on a first virtual machine. The method may also include determining a first communication path between the process included in the container on the first virtual machine and a hypervisor of the virtualization system, and migrating the container and the process included in the container to a second virtual machine. The second virtual machine may be distinct from the first virtual machine. Additionally, the method may include determining a second communication path between the process included in the migrated container on the second virtual machine and the hypervisor of the virtualization system.