Method for distributing resource in large scale storage system
A technology for resource allocation and mass storage, applied in memory systems, transmission systems, instruments, etc., to solve problems such as failure to optimize storage system performance and disk damage.
Patent Information
- Authority / Receiving Office
- CN · China
- Current Assignee / Owner
- Publication Date
- 2007-05-02
- Estimated Expiration
- Not applicable · inactive patent
Smart Images
Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The resource allocation method in the mass storage system belongs to the field of storage area network, especially relates to the field of data distribution and load balancing therein. Background technique
[0002] With the rapid development of information technology, the amount of data that people have to deal with continues to increase; this puts forward huge demands on the capacity and performance of storage systems, and makes traditional storage methods unable to meet the needs of users. A storage area network (Storage Area Network, SAN) is a specialized storage network, which can connect various storage devices together to form a unified management of storage resources, and hosts can access these resources through optical fiber or IP networks. At the same time, in order to support disks of different specifications and various heterogeneous front-end hosts in the storage network, virtualization technology is introduced into the storage area network...
Examples
Embodiment Construction
[0070] The core of the present invention is to maintain a set of storage resource configuration information through dedicated virtualization processing nodes to centrally manage various storage device resources in the storage network and perform resource allocation. The virtualization processing node is a server running a Linux operating system. At the same time, it provides completely transparent storage services to the front-end host through the SCSI middle layer for command analysis and processing. The specific implementation architecture includes the following two main parts:
[0071] 1. The configuration information list of all storage devices (including physical devices and virtual devices) is maintained on the virtualization processing node: for each PD, record its various attributes and space lists, that is, the SG table; Resource allocation, and create a structure for each VD, record various attributes and space mapping methods.