Automatic simplified-configured self-adaptation optimization method

Abstract

The invention discloses an automatic simplified-configured self-adaptation optimization method which comprises the following steps that S1. during initial running of a storing system, data blocks with fixed size are used for simplified distribution; S2. during running, the feature of a data IO request which is sent to the storing system is subjected to statistics; S3. according to the IO request statistics results, the size of the data blocks which are subjected to simplified distribution is adjusted automatically; S4. during system running, the space using situation of large data blocks is recorded, and meanwhile data migration and data block merging are carried out; and S5. when the storing system receives a space recycling request, recyclable data blocks are recycled, and data block merging is carried out. The distribution size of the data blocks can be adjusted automatically according to the storing system practical service requirement, on the basis that the requirement of management according to needs of a disk space is met, extra storing system performance expenditure caused by data block distribution is lowered, the disk space using rate is improved, and accordingly the storing system is suitable for various service application scenarios.

Description

technical field [0001] The invention relates to the field of storage technology, in particular to an adaptive optimization method for thin provisioning. Background technique [0002] Thin provisioning technology is a new storage space configuration management technology. Compared with traditional storage space management technology, thin provisioning technology can allocate storage resources on demand according to the actual needs of upper-layer applications for storage space, rather than one-time Allocate all storage resources required by the application. Thin provisioning technology solves the problem of low storage space utilization in traditional storage systems, improves storage space utilization, and reduces storage system deployment and operation costs. [0003] Thin provisioning can be implemented at different levels of granularity such as logical volumes, files, and data blocks. At present, most storage systems use thin provisioning at the data block level to divid...

AI Technical Summary

Problems solved by technology

[0004] Compared with allocating the required storage space for logical volumes at one time, the thin provisioning technology dynamically allocates storage resources. The resulting impact is that the process of a disk IO issued by the upper-layer application increases the steps of space allocation. , increases the IO path, and at the same time, due to the additional maintenance of metadata for data block management, it will have a certain impact on the data access performance of the storage system

Therefore, the choice of data block size in thin provisioning implementation will affect the overall performance of the storage system and the optimization degree of storage space utilization

When the data block is large, less metadata is required, and the performance overhead caused by data block allocation is small when the storage system is running, but the improvement in storage space utilization is small. At the same time, large data will be generated during disk space recovery. The problem of low space utilization in the block; when the data block is too small, the storage space utilization will be greatly improved, but a large amount of additional metadata information is required, and at the same time, the data block allocation overhead is large when the storage system is running normally

Images