VM Live Migration Flow Caching for Faster TCP Session Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional virtual machine live migration technologies suffer from slow network flow restoration of TCP sessions, leading to prolonged downtime and poor user experience due to the suspension of the source virtual machine during memory copying.
Innovation Solution
Implement a flow cache management module on the destination physical machine to cache TCP data packets sent to the suspended virtual machine, which are then delivered upon restoration, thereby accelerating the TCP connection restoration process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If the source virtual machine is suspended during memory copying to ensure data consistency, then the accuracy of memory copying is improved, but the downtime of the virtual machine increases
Solution Approach 1:
The patent applies preliminary action by establishing a flow cache mechanism before the virtual machine suspension. The flow cache pre-prepares to store incoming data packets, so when the VM is suspended and memory is being copied, network traffic is not lost but temporarily held. This allows the VM to be suspended for the necessary time to ensure memory copying accuracy without permanently losing network connectivity, as the cached packets can be delivered after resumption.
2Stability of the object's composition
If the virtual machine is suspended for final memory copying, then the consistency of virtual machine state is improved, but the network connection restoration time increases
Solution Approach 1:
The flow cache acts as an intermediary between the network and the virtual machine during suspension. Instead of directly losing packets when the VM is suspended, the flow cache mediates by temporarily storing incoming packets and forwarding them after the VM resumes. This intermediary mechanism decouples the network traffic flow from the VM's operational state, allowing the VM to be suspended for state consistency without directly impacting network connection restoration time.
3Reliability
If TCP timeout retransmission is used to restore connections after migration, then the reliability of connection restoration is improved, but the time to restore TCP connection increases
Solution Approach 1:
The flow cache performs preliminary action by pre-storing incoming data packets before the VM suspension completes and before TCP timeout retransmission would be triggered. When the VM resumes, the cached packets are immediately available for delivery, which can prevent or reduce the need for TCP timeout retransmission. This preliminary caching action maintains connection reliability while significantly reducing the time penalty that would otherwise be incurred through TCP timeout mechanisms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for live migration of a virtual machine includes receiving a data packet that is sent to a migrated virtual machine on the source physical machine in a stage when the migrated virtual machine is suspended, and caching the received data packet; and sending the cached data packet to the migrated virtual machine on the destination physical machine after it is sensed that the migrated virtual machine is restored at the destination, to speed up restoration of a TCP connection inside the virtual machine. The apparatus of the present disclosure includes a caching unit and a data restoration unit. The method and apparatus of the present disclosure improve a restoration speed of the TCP connection, make live migration of a virtual machine more imperceptible for users, and improve user experience.