Cloud Aqua: Object-based Cloud Storage

With the arrival of big data era, conventional storage is barely able to meet the requirements of real time dynamic management, transfer, sharing and application of big data. Cloud computing and cloud storage have begun providing the solutions and development space for the management and application of big data. Establishment of object’s data service based on the framework of cloud storage is the best solution for the storage and application of massive big data from data center.

XOR’s Cloud Aqua, the object’s data storage service, is multi-site based shared storage service to realize remotely or cross network real time sharing and intelligent management of big data file, virtualization and consolidation of storage resources, and making management more effectively. Storage space and network resources are allocated by actual needs. The allocation quota can be adjusted flexibly by the change of business needs for enhancing utilization and efficiency. Storage can be upgraded to the order of magnitude of PB. It can achieve a higher level of data security and disaster recovery, quicker and easier data recovery.

CloudAqua’s key functional framework unifies management (such as storing, copy, disaster recovery, transfer, security, etc.) of storage object’s files across physical locations and networks by using multiple engines. Designed as distributed, its engines can easily expand the services at any time as needed possessing greater flexibility.

Private Cloud Storage Platform
Across locations and networks
- Universal storage pool. Support Heterogeneous devices.
- Multi-tenancy. Allocate recourses on demand
- Auto-management on big data
  • distributed object storage
    - higher elasticity
    - System scales on demand
    - I/O performance improves as the number of storage nodes increases

    WAN Acceleration
    - UDP protocol based peer-to-peer(P2P) for unicast/multicast, speed up file transfer, and consume less network resources
    Enriched Interfaces
    - Various SDK package interfaces (javascript, dotnet, etc) and plug-ins of mainstream OS
    - NAS and iSCSI
    - RESTFUL API (CDMI and S3)
    Improved Management and Processing Capability
    Outstanding Metadata Management
    - Metadata management based on NoSQL in a federated structure
    - Quick scan and management
    - Metadata database can be used as appliance database
    Empowered Storage Ability
    - Software-defined Storage
    - Support various storage platforms and tools, including Ceph, XOR SHAS and LVM

    Intelligent Policy Engines
    - Automated data management based on pre-defined policies
    - Policy is triggered by message, and realized through queues. Safe and quick
    Big Data Analysis
    - Massive big data queries accomplished by means of object’s metadata
    - Able to work collaboratively with Map-Reduce

    Optimized for digital media
    - Broadcast-level multi-media
    - Supports various media processing including Capture, analyze, and search Transcoding, voice recognition, and facial recognition
    convenient development platform
    - Modeling in client applications can be mapped directly to CloudAqua’s object model
    - Specialized functions, small granularity, easy to integrate
    Higher security and fault-resilience
    Storage Safety
    - Replications ensuring higher security
    - Soft-raid data protection
    Access Safety
    - Integration with LADP and Active Directory
    - Folder level access control
    - Audit log records all operations to the system and files

    - Copy on write
    -Snapshots generating, viewing, rolling back, restoring and deleting
    Fault resilience
    - Failover among multiple sites to realize Fault resiliency

    Straightforward Monitoring and Managing

    Straightforward Monitoring and Managing through user friendly web interface

    Architecture: Cloud Aqua Storage Architecture and Interfaces

    Scene 1: Real Time Sharing Across Locations and Networks

    Scene 2: Message Queues Facilitating Applications Collaboration