Shulou(Shulou.com)06/01 Report--

The main content of this article is to explain "what is the difference between DAS/NAS/IP SAN/FC SAN". Interested friends may wish to have a look. The method introduced in this paper is simple, fast and practical. Now let the editor take you to learn "what is the difference between DAS/NAS/IP SAN/FC SAN"?

DAS: server directly rear-mounted storage device, the most economical structure.

NAS: the storage device directly attached to the network is actually a file sharing server on Ethernet using NFS, CIFS and other network file systems.

SAN is a disk on a network, and NAS is a file system on a network.

IP SAN: SAN (storage area network) network using iSCSI technology, and the transmission medium is IP network.

FC SAN: is the application of optical fiber technology SAN network, the transmission medium is optical fiber, the highest performance, currently the most widely used.

1. Directly connected storage (Direct Attached Storage-DAS)

The storage device is connected directly to the server via a cable (usually a SCSI interface cable). The Icano request is sent directly to the storage device. This is done by connecting separate or two small cluster servers.

Its characteristic is that the initial cost may be relatively low. However, under this connection mode, for the environment of multiple servers or multiple PC, each PC or server has its own storage disk, so it is difficult to redistribute the capacity; for the storage system management in the whole environment, the work is tedious and repetitive, and there is no centralized management solution. Therefore, the overall management cost is higher.

Disks in PC or JBOD with only one external SCSI interface are part of the DAS schema.

two。 Network attached storage (Network Attached Storage-NAS)

NAS devices are usually integrated with processors and disk / disk enclosures, similar to file servers. Connect to the TCP/IP network (through LAN or WAN) and access data through file access protocols such as NFS,CIFS, etc.

NAS translates a file access request into an internal Imax O request. This way is to connect the storage device to the IP-based network, unlike DAS and SAN, the server sends a file access request to the storage device NAS through the "File ID O" mode. NAS is generally installed with its own operating system, which converts File Imax O to Block Imax O and sends it to the internal disk.

NAS system has low cost and is easy to realize file sharing. However, because it uses the way of file request, the performance of the device requested by the block is poor, and the NAS system is not suitable for systems that do not use the file system for storage management, such as some databases.

NAS must have access to the questionnaire or physical disk. NAS must have the ability to access Ethernet, that is, it must have an Ethernet card.

3.IP SAN

If SAN is a TCP/IP-based network, implement the IP-SAN network. In this way, the server and the storage device are connected through a dedicated network, and the server sends a data access request to the storage device through the "Block Ithumb O".

The most commonly used technology is iSCSI, which transmits SCSI command packets in TCP/IP packets, that is, SCSI over TCP/IP.

The advantages of IP SAN are:

The ubiquitous Ethernet network protects existing investments to some extent.

IP storage goes beyond the limit of geographical distance and is suitable for remote backup of critical data.

IP network technology is mature, there is no interoperability problem

IP storage reduces the complexity of configuration, maintenance and management.

IP network has been widely recognized by the IT industry-network management software and service products are available

The wide use of gigabit network has greatly improved the performance of IP network.

With the development of 10 Gigabit network technology, IP storage can surpass FC storage in performance.

4. Storage area Network (Storage Area Network-FC SAN)

Storage devices form a separate network, mostly connected by optical fiber, using fibre Channel protocol (Fiber Channel, referred to as FC). Any connection can be made between the server and the storage device, and the Imax O request is also sent directly to the storage device.

Fibre Channel protocol actually solves the underlying transport protocol, while the high-level protocol still uses SCSI protocol, so fibre Channel protocol can actually be regarded as SCSI over FC.

The advantages of storage area networks are as follows:

Greater distance between the server and the storage device (fibre Channel network: 10 km compared to DAS's SCSI:25 meters)

High reliability and high performance

Multiple servers and storage devices can be connected at will

Centralized storage devices replace multiple independent storage devices to support storage capacity sharing

Through the corresponding software, the storage devices on SAN are shown as a whole, so they have high scalability.

Storage devices on SAN can be centrally managed and controlled by software to provide data sharing

Because SAN is usually a fibre Channel-based solution, it requires dedicated fibre Channel switches and management software, and the initial cost of SAN is higher than that of DAS and NAS.

Comparison of SAN and NAS:

1. NAS devices can generally provide multiple protocols to access data, while SAN can only use SCSI protocol to access data.

2. NAS can realize shared access by multiple clients on a disk array, including accessing a directory or file at the same time. In SAN mode, unless all clients have installed special cluster management system or cluster file system module, a LUN cannot be shared, and forced sharing will destroy the data.

Performance:

FC has the best performance, and a single port can reach the bandwidth of 2Gbps; NAS has the worst performance, even if it uses a gigabit network, it is usually only 30MB / 40MBps; a single gigabit NAS can reach 60mur70MBps; under a 10 Gigabit network, the speed of a single 10 Gigabit port can exceed 500MBps.

Scalability:

FC and iSCSI adopt the architecture of SAN and have the best scalability. In the storage network, it is easy to add users or storage modules.

Application cost:

FC is the most expensive application, requiring an expensive 8-port fiber switch (30-40, 000 yuan), a fiber channel card (50, 000-10, 000 yuan) per server, and fiber-optic media; by contrast, iSCSI only needs an ordinary Ethernet switch (1000 yuan for an 8-port gigabit switch) The server itself has a gigabit network port, even if an additional network card is 500 yuan, you can use ordinary Category 5 lines, and the price is negligible.

Market positioning:

FC occupies the high end, iSCSI is in the middle and high end, NAS and DAS are in the low end.

As more and more organizations transform the data center infrastructure, most customers know more about SAN and begin to apply it. But recently, many manufacturers have begun to introduce IP SAN to their customers, which makes many customers have a lot of doubts about what is FC SAN, what is IP SAN, the difference between the two and how to choose in their own construction environment. This article gives the corresponding explanation on the characteristics, differences, maturity, and how to choose between FC SAN and IP SAN, for your reference.

Project

IP SAN

FC SAN

Description

Technology maturity

IP SAN is a new technology that has been implemented for the business and known to the users in the last 2 years.

FC SAN is a storage network technology that has been developed since the late 1990s, and its development has gone through at least three generations: 1Gb, 2Gb and the current 4Gb. It has been used for nearly a decade, and it is a very mature and reliable technology, and its adoption and recognition can be said to be widespread in the data centers of almost all kinds of institutions, large and medium-sized enterprises.

In terms of technology maturity, FC SAN is much higher than IP SAN

Agreement efficiency

IP SAN essentially encapsulates SCSI instructions in IP packets and transmits packets using IP technology, taking advantage of the universality and popularity of IP technology. But IP encapsulation has a significant weakness: the overhead and inefficiency of IP encapsulation-- that is, the header and trailer to be appended to any IP packet, and too many verification codes, so its overall efficiency is not high.

Compared with FC SAN, the SCSI instruction is encapsulated in the FC package. The proportion of header, trailer and check code is very low, so its efficiency is very high.

In terms of efficiency, FC SAN is significantly higher than IP SAN, so FC SAN is more suitable for efficiency-sensitive applications, such as database applications with high performance requirements, while IP SAN is mainly used in environments that do not require high performance and efficiency, such as OA, document processing, multimedia environment, etc.

Performance

The encapsulation efficiency of IP SAN protocol is not high, so IP SAN requires better hardware speed of the environment in order to achieve the same performance as FC. Unfortunately, at present, the best environment of IP SAN is only in gigabit network, so its performance can not be compared with FC SAN.

The FC SAN protocol itself is efficient, and at present, FC SAN has begun to deploy 4Gbps environment generally, so FC SAN has a lot more performance blocks than the current IP SAN.

Optimistically, the performance of 10Gb, that is, IP SAN in 10 Gigabit network, may be significantly improved, and can initially meet the requirements of related applications, which can be compared with the performance of FC SAN at present.

Stability and security

Low

High

IP SAN is built on the ordinary IP network, FC SAN is built on the FC network. The anti-interference ability of FC network is strong.

At the same time, the closeness of the FC network is higher, do not want the IP network to be very open, so the FC SAN protocol should be relatively secure and stable.

Cost

Low

High

The low cost of IP SAN is mainly reflected in:

-low price of equipment

-low cost of operation and maintenance

Suggestion

From the above analysis, we suggest that:

-in non-critical environments, under the condition of very limited construction cost, IP SAN can be considered.

-for the core business and key business of an enterprise or organization, which requires high stability, performance and technical maturity, it is recommended to adopt FC SAN, so that the main business system of the enterprise and organization will be more guaranteed.

Other aspects

Compatibility: at present, IP SAN mainly completes the compatibility testing of lower-end servers such as Windows,Linux.

Vendor support:

-in terms of servers, it is explicitly supported by PC Server manufacturers and low-end Unix servers, but not by some high-end servers.

-Storage: although most storage supports IP SAN, the main application in the user environment is low-and middle-end storage.

FC SAN compatibility testing is very sufficient and can be supported by all high-end, middle-end, and low-end servers.

Vendor support:

FC is fully supported on almost all servers (regardless of class) and stand-alone storage systems, regardless of server or storage

In terms of compatibility and vendor support, it is also recommended that FC should be more stable and reliable for important and critical business systems.

FC SAN and IPSAN instructions:

The birth of SAN (full name Storage Area Network, storage local area network) makes the storage space more fully utilized, and makes the installation and management more effective.

SAN is a technology that integrates storage devices, connecting devices, and interfaces in a high-speed network. SAN itself is a storage network, which undertakes the task of data storage. The SAN network is isolated from the LAN service network, and the storage data stream will not occupy the business network bandwidth.

In SAN network, all data transmission is carried out in high-speed and high-bandwidth network. SAN storage realizes block-level storage access to physical hardware directly, which improves storage performance and upgrade ability.

The early SAN uses fibre Channel (FC,Fiber Channel) technology, so the previous SAN mostly refers to the storage local area network using fibre Channel. After the emergence of the iSCSI protocol, in order to distinguish, the industry divided SAN into FC SAN and IP SAN.

ISCSI (Internet small computer system Interface) is a standard for data block transmission over TCP/IP. It is initiated by Cisco and IBM, and has been strongly supported by major storage manufacturers. ISCSI can run SCSI protocol on IP network, so that it can perform fast data access and backup operations on high-speed Gigabit Ethernet.

The iSCSI standard was certified by IETF (Internet Engineering Task Force) on February 11, 2003. ISCSI inherits the two most traditional technologies: SCSI and TCP/IP. This has laid a solid foundation for the development of iSCSI.

ISCSI-based storage systems can achieve SAN storage functions with little investment, or even directly make use of the existing TCP/IP network. Compared with the previous network storage technology, it solves the problems of openness, capacity, transmission speed, compatibility, security and so on.

In practice, SCSI commands and data are encapsulated in TCP/IP packets and then transmitted over the IP network. The specific workflow is as follows:

(1) the iSCSI system sends a SCSI command by the SCSI adapter.

(2) the command is encapsulated in a TCP/IP packet and sent to the Ethernet network.

(3) the receiver extracts SCSI commands from the TCP/IP packet and performs related operations.

(4) encapsulate the returned SCSI commands and data into TCP/IP packets and send them back to the sender.

(5) the system extracts data or commands and sends them back to the SCSI subsystem.

At this point, I believe you have a deeper understanding of "what is the difference between DAS/NAS/IP SAN/FC SAN". You might as well do it in practice. Here is the website, more related content can enter the relevant channels to inquire, follow us, continue to learn!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.