Product Introduction
The DS-A5000 series is a storage product that combines new AI intelligent technology, big data technology, and cloud computing technology on the basis of video cloud storage. Through system optimization and integration, it provides a series of functions including storage, calculation, intelligent analysis, big data query, and platform application on a storage platform architecture. It integrates intelligent algorithms based on deep learning and can provide recognition, analysis, modeling, comparison, query, alarm, and other functions for facial, vehicle, and human behavior;
A large-scale horizontally scalable storage system designed for the Internet of Things, cloud computing, and big data. By providing a reliable, efficient, and elastic new storage service platform and adopting an open architecture of resource sharing, on-demand expansion, and business integration, it helps users quickly achieve "digital transformation" and easily meet the large-scale digital agile access needs in the Internet of Things environment.

Product characteristics

openness

Standard protocols: Provide standard protocols such as stream direct storage, ISCSI, NFS, CIFS, S3, Swift, etc., seamlessly integrating with various business systems;
Virtual Platform: Provides KVM and Docker standard virtual computing resources
Open docking: providing SDK and API secondary development packages

fusion
AI powered: GPU virtualization, AI algorithm warehouse, big data comparison, etc
Storage computing: seamless integration of storage and computing, elastic scalability;
Video application: Built in streaming storage, streaming media and other video software, embedded with intelligent application platform applications;

flexible
On demand configuration: providing various disk positions, GPU computing power, and software matching chips to meet different software and hardware configurations according to needs;
Elastic and efficient: No need for detailed planning, elastic expansion, efficient utilization
Open compatibility: compatible with various standard storage interfaces and algorithm packages

reliable
Data protection: adopting multiple replicas and N+M: K redundancy mechanism, allowing multiple disks or nodes to fail without losing data;
Cluster deployment: Supports single machine cluster deployment with multi-level software and hardware protection
Business Balance: Provide cluster HA mode for business provision, load balancing

Functionality

Fully symmetrical distribution：
Uniform distribution of data, eliminating metadata access bottlenecks, providing load balancing, and ensuring system performance under scale expansion mode;
Elastic and efficient：
Support single node deployment, seamless scale out smooth expansion of the system, linear increase in capacity and performance;
Multiple interfaces：
Support stream direct storage iscsi、FC、 Multiple access interfaces such as blocks, objects, and files, supporting seamless integration with HDFS and RBD;
Rich variety of models：
There are multiple disk slots to choose from, supporting various software and hardware matching modes such as P4, T4, AI embedded smart card, etc; ?
uniform naming：
Adopting a unified IP and namespace, simplifying system management and maintenance, and eliminating data silos caused by multiple naming conventions;
Flexible Matching：
Software and hardware are available on demand, with different hardware and software modules used for different scenarios, saving users investment;
AI Features：
Built in comprehensive analysis software, supporting algorithm repositories, various algorithm packages, big data queries, and peer-to-peer alarm functions;
Human centered operation and maintenance：
Provide configuration wizard, one click configuration and other functions, allowing users to quickly complete system configuration; Provide rich alarm management methods to improve equipment maintenance efficiency;