About Data Storage Systems

A mass amount of data is annually generated and stored in IT corporations, banks, insurance companies, telecommunication industries, and government and military organizations. However, the traditional storage mechanisms cannot provide the ever increasing dependability and performance needs of users, mandating the use of data storage systems. These systems, classified to low-end, mid-range, and high-end, offer a high level of availability, reliability, and performance, and support special features such as snapshot, instant copy, and remote backup.

Data Storage Methods

A variety of data storage methods can be noted such as Direct Attached Storage (DAS), Network Attached Storage (NAS), and Storage Area Network (SAN). Each method is preferred depending on the users’ requirements.

DAS
In DAS structure, a direct connection provided between servers and the data storage system. This structure is preferred for local data access and sharing data in the environments with a few number of servers. In DAS structure the storage device is attached to server by parallel or serial ports, while these ports can support a limited distance of storage device and server.
NAS
NAS structure is used in file servers. This structure connects storage to a LAN using IP, and enables file access to clients. In this structure, the exact physical location of data storage is transparent in the user side, as it is managed by the file system of NAS host operating system. This structure usually offers a higher level of scalability, availability, and performance, compared to general purpose file servers, due to its special design optimized for file management.
SAN
SAN is now recognized as the most efficient storage structure, offering block-level access using high-speed connections such as Fiber Channel, iSCSI, and InfiniBand between servers and storage system. This structure provides higher scalability, availability, performance, and cost efficiency compared to DAS structure. SAN storage systems support high performance, high security, large distance between storage and servers, and easy backup facilities compared to other storage structures, while having higher complexity and cost.
Storage Structure Network Connection Media Input/Output Protocols Bandwidth Availability
DAS No Parallel SCSI / FC SCSI 40 MBps up to 160 MBps Low
NAS Yes Ethernet NFS/CIFS 10 MBps to 1 Gbps Moderate
SAN Yes Fiber Channel/Ethenet SCSI 1Gbps up to 50Gbps High

SAB storage system has SAN structure, using Fiber Channel switches and cables for the connection of servers and storage arrays. This storage system uses a dedicated software/OS for storage management. The following figure shows the structure of SAB storage system and its connection to servers.

Main features of Data Storage System

The continuous service of data centers is of decisive importance, mandating high reliability and availability of data storage system. The main features of a SAN storage system are as follows:

High Availability
Availability is defined as the probability that the storage system is operating in a true condition and the data is available to the users. High availability is of major features of storage systems, as the data unavailability can seriously impact businesses and corporations (Learn More).
High Capacity
The storage needs may change in time. Hence, the maximum supported capacity of a storage system is of decisive importance. The storage system should support hot scalability to increase storage capacity without service interrupt and downtime (Learn More).
High Scalability
Two metrics, throughput and response time, are prevalently used to measure the performance of a data storage system. Throughput stands for the number of input/output operations the storage systems can handle in each second, and response time is the average time between a request is received and the time it is responded by data storage system. Higher throughput and lower response time is preferred (Learn More).
High Performance
Two metrics, throughput and response time, are prevalently used to measure the performance of a data storage system. Throughput stands for the number of input/output operations the storage systems can handle in each second, and response time is the average time between a request is received and the time it is responded by data storage system. Higher throughput and lower response time is preferred More Details
Data Integrity
Data Integrity refers to the maintenance of data and assurance of data accuracy and consistency within mission time. Decreasing the chance of human errors, by decreasing the role of human agents in maintenance and service tasks and improving the service/maintenance mechanisms, can increase data integrity.
Security
Security refers to all policies and methods preventing unauthorized data access in a data storage system.
Contact Info
No. 3, Golestan Alley, Shahid Ghasemi St, Shahid Akbari Blvd, Tehran, Iran
+98 (21) 66029164
info@hpds.ir

Support