Big data storage solution

1. Overview of requirements

As the internal data of the enterprise continues to grow, enterprises will find that the original storage architecture is difficult to meet the corresponding storage and management needs. According to a recent IDC survey of online storage services (SaaS), online storage services will grow faster than traditional storage architectures in the next few years due to the inability of IT departments to access more expenses and technicians. speed. Online storage services are a form of service based on the big data storage architecture.

2. the solution

Compared with traditional storage devices, big data storage is more than just a piece of hardware, but a complex system composed of multiple parts such as network devices, storage devices, servers, application software, public access interfaces, access networks, and client programs. . Each part is based on a storage device and provides data storage and service access services through application software.

The structural model of the UIT big data storage system consists of four layers.

First, the storage layer

The storage layer is the most basic part of big data storage. Above the storage device is a unified storage device management system, which can implement logical virtualization management of storage devices, multi-link redundancy management, and status monitoring and fault maintenance of hardware devices.

UIT adopts the object storage system architecture to realize the cluster work of the back-end storage devices through the object storage cluster file system, and realizes the management of the entire system, data distribution, processing, and processing structure feedback through the system control unit and management unit. At the same time, through various data backup and disaster recovery technologies and measures, data in big data storage will not be lost, and the security and stability of big data storage itself will be guaranteed.

Second, the application interface layer

The application interface layer is the core of big data storage and the most difficult part of big data storage. The application interface layer implements the collaborative work between multiple storage devices in the big data storage through technologies such as the cluster, the distributed file system, and the grid computing, so that multiple storage devices can provide the same service to the outside, and provide more and more. Stronger data access performance.

Third, the application software layer

The application software layer is the most flexible part of big data storage. Different big data storage operating units can develop different application service interfaces according to actual business types and provide different application services. For example, data remote disaster recovery and remote backup, video surveillance application platform, IPTV and video on demand application platform, network hard disk reference platform, remote data backup application platform.

Fourth, the access layer

Any authorized user can log in to the big data storage system through the standard public application interface and enjoy the big data storage service. Big data storage operating units are different, and the access types and access methods provided by big data storage are also different. Big data storage users use different application software clients and enjoy different types of services.

3. functional advantages

1, cluster function

Ability to obtain PB level storage capacity through cluster
Support for general purpose servers, general purpose storage devices

Add and delete storage nodes online without interrupting the entire cloud

2, performance

Concurrent storage of individual files
Automatic load balancing of device capacity

Automatic load balancing for hotspot files

3. Reliability

Data redundancy and backup

Automatic fault handling

4, management

Based on object-based file management, files are hierarchically managed through policies such as reliability and accessibility.
Automatic configuration, no additional manual configuration work by the administrator after software installation.
Automatic fault detection and recovery
A web-based interface is required to allow administrators to monitor and adjust the operation of the storage cluster

Basic control over user access rights and space

5, visit

Global unified namespace
Provides a standard NAS access interface
Provide efficient concurrent access protocols (or APIs)
Web access interface based on HTTP/REST API
Customer Service Hotline

© 2016 United Information Technology Co., Ltd.粤ICP备05121547

Legal | Privacy | Sitemap | Careers | Feedback | Contact Us