Online migration redundancy removal method for virtual machine disk

Abstract

The invention discloses an online migration redundancy removal method for a virtual machine disk. The method comprises the following steps: firstly, reading a super block in a disk according to a filesystem organization mode, then reading a bitmap in each block group according to block group description information in the super block, determining whether each block in the bitmap record block group is used or available, and constructing a complete bitmap for the whole disk by utilizing the obtained bitmaps of all the block groups; secondly, in the virtual machine migration process, carrying out disk migration according to the obtained disk use information, and at the moment, only transmitting used disk blocks; and thirdly, extracting unused block bitmaps which are possibly updated from theoriginal memory data by using a virtual machine introspection technology, updating bitmap information by using the unused block bitmaps after obtaining caches of metadata of all block equipment in the memory, and migrating used disk blocks which are not migrated before to a destination end by using an existing method. According to the method, semantic information of the data is mainly used for data compression, so that very small overhead for upper-layer applications is introduced.

Description

technical field [0001] The invention belongs to the field of data center management, and in particular relates to a virtual machine disk online migration and de-redundancy method. Background technique [0002] Virtual machine migration technology mainly refers to the transfer of a virtual machine and its running status from one computing node to another, ensuring that the virtual machine can continue to provide services at the destination when the source computing node is shut down due to failure or maintenance. Traditional virtual machine migration mainly assumes that virtual machines are migrated in the same data center or cluster. However, in order to meet the growing demand for large-scale application deployment and higher reliability requirements, the demand for cross-WAN migration is getting higher and higher. Unlike data center migration, WAN migration requires disk migration. However, the disk data volume is often large and consumes a lot of network bandwidth. [00...

AI Technical Summary

Problems solved by technology

[0009] 1. The compression algorithm used in the above method will consume a lot of CPU resources, which will have a very large performance impact on the virtual machine migration due to resource constraints; in addition, indexing will also consume a lot of storage space

[0010] 2. The above method fails to effectively utilize the upper-level semantic information of the data during the compression process, so the compression algorithm is blind, resulting in a large waste of resources

Images