SAN, NAS, DAS, and sannasdas

SAN, NAS, DAS, and sannasdas

※Today, I am free to sort out the SAN, NAS, and DAS-related things and share them with you. If there is any error, I hope you will forgive me and give me more corrections.

In network storage, there are various network storage solutions, such as SAN, NAS, and DAS storage networks. They have their own characteristics and their application scenarios are also different. The following describes their respective features.

I. SAN

A Storage Area Network (SAN) is a high-speed, dedicated Network for Storage operations. It is generally independent of a computer LAN ). SAN connects hosts (Management Servers, business servers, etc.) with storage devices to provide dedicated communication channels for any host and any storage device. SAN isolates storage devices from servers to share storage resources at the server level. SAN introduces channel and network technologies into the storage environment and provides a new network storage solution that can meet throughput, availability, reliability, scalability, and manageability requirements.

Typically, a SAN consists of a disk array (RAID) connected Fiber Channel (to distinguish it from an ip san, also known as a FC-SAN ). SAN and server and client data communication uses SCSI commands instead of TCP/IP. Data Processing is block level ). SAN can also be defined as a data storage center. It adopts a scalable network topology and uses a direct connection mode with high transmission rates through optical channels, provides multiple data exchanges between any node in the SAN, and centrally manages data storage in a relatively independent storage area network. SAN will achieve maximum data sharing and Optimized Data Management under multiple operating systems, as well as seamless system expansion.

Among them, SAN network is subdivided into FC-SAN network and IP-SAN network.

1. FC-SAN

FC-SAN, as its name suggests, is to directly connect to the disk array through the FC channel, data directly communicates with the hardware by sending SCSI commands, thus improving the overall rate.

The composition of the FC-SAN:

There are some dedicated hardware and software in the FC-SAN. The Hardware includes the FC card, fc hub, FC switch, and storage system. The software is mainly the FC controller driver and storage management software for various operating systems.

① FC card: it is mainly used for the connection between the host and the FC device.

② Fc hub: the internal operation of the arbitration ring topology, the nodes connected to the HUB share 100 MB/S band width (or higher ).

③ FC switch: the Fabric topology is run internally. Each port occupies an exclusive 100 MB/S band width (or higher ).

④ FC storage device: one or more FC interfaces can be connected. FC storage devices usually use fiber-optic hard disks, as well as fiber-to-SCSI solutions. Using SCSI (or ATA) hard disks, the entire configuration is cheaper.

⑤ Storage network management software: the main function of the storage management software is to automatically discover the network topology and ing. It is automatically discovered and configured when the storage network is increased or decreased.

High Performance Fiber Channel switch and fiber channel network protocol is the key of FC-SAN. The network topology with the fiber channel switch as the backbone is called "SAN Fabric ". Fiber Channel Protocol is another essential feature of FC-SAN. FC-SAN is the use of fiber channel protocol attached to the SCSI protocol to achieve reliable block-level data transmission.

Application scenarios of FC-SAN:

Since FC-SAN is optimized to transfer large volumes of data between servers and storage devices, it is ideal for the following applications:

① Key task database applications. The predictable response time, availability, and scalability are the basic elements.

② Centralized storage and backup, in which performance, data consistency and reliability can ensure the security of key enterprise data.

③ High Availability and Failover environments can ensure lower costs and higher application levels.

④ Scalable Storage Virtualization separates storage from direct host connections and ensures dynamic storage partitions.

⑤ The improved Disaster Tolerance feature provides fiber channel performance and extended distance between host servers and Their connected devices.

Advantages of FC-SAN:

In the face of rapid growth in data storage demand, enterprises and service providers are gradually choosing FC-SAN as the network infrastructure, because SAN has excellent scalability. In fact, SAN has more significant advantages than traditional storage architecture. For example, traditional server connection storage is usually difficult to update or centrally manage. Each server must be shut down to add and configure new storage. In comparison, the FC-SAN does not have to go down and interrupt connections to the server to increase storage. FC-SAN can also centrally manage data, reducing total cost of ownership.

Using Fiber Channel Technology, FC-SAN can effectively transmit data blocks. SAN provides an effective method for data backup by transferring massive data blocks between storage and servers. Therefore, the network bandwidth traditionally used for data backup can be saved for other applications.

Open, industry-standard fiber channel technology also makes FC-SAN very flexible. FC-SAN overcome the limitations of traditional cables connected to SCSI, greatly expanding the distance between servers and storage, thus increasing the possibility of more connections. The improved scalability also simplifies the deployment and upgrade of servers and protects the investment of original hardware devices.

In addition, FC-SAN can better control the storage network environment, suitable for those trading-based systems in terms of performance and availability needs. SAN uses high-reliability and high-performance fiber channel protocols to meet such requirements.

Another advantage of FC-SAN is the ability to transfer data blocks to enterprise-class data-intensive applications. In the process of data transfer, FC-SAN processing costs less on the Communication Node (especially the server), because the data is divided into smaller data blocks during transmission. Therefore, fiber channel FC-SAN is very effective when delivering big data blocks, which makes the Fiber Channel Protocol very suitable for storage-intensive environments.

2. IP-SAN

Simply put, the IP-SAN (IP Storage) communication channel uses an IP channel, instead of a fiber channel, to connect the server to the storage device technology, in addition to the standard iSCSI, there are also FCIP, iFCP and other standards being developed. The fastest growing iSCSI has become a powerful representative of IP Storage.

Like fiber channel, IP Storage is interchangeable, but unlike Fiber Channel, IP networks are mature and there is no interoperability problem. This is the biggest headache for Fiber Channel SAN. IP has been widely recognized by the IT industry and many network management software and service products are available for use.

Ii. NAS

Network Attached Storage (NAS. In the NAS storage structure, the storage system no longer attaches to a server or client through the I/O bus, but is directly connected to the network through the network interface and accessed by the user through the network.

NAS is actually a storage device with a thin server. Its function is similar to a dedicated file server. This type of dedicated storage server removes most of the computing functions that are not applicable to the general server, and only provides file system functions. Compared with traditional server-centered storage systems, data is no longer forwarded through the server memory and directly transmitted between the client and the storage device. The server only plays a role in control management.

NAS features:

NAS uses the traditional Ethernet protocol. During file sharing, NFS and CIFS are used to communicate with NT and Unix systems. Because both NFS and CIFS are based on the file sharing protocol of the operating system, NAS Features shared access to small files.

NAS devices are directly connected to Ethernet memory and provide file services to clients using standard network file systems such as NFS and SMB/CIFS over TCP/IP interfaces. NAS devices provide file-level services to clients. However, it still communicates with its storage devices at the data block level. The file system is in this NAS storage.

NAS advantages:

NAS is applicable to users who need to transmit file data to multiple clients over the network. NAS devices can play a good role in environments where data must be transmitted over long distances.

NAS devices are very easy to deploy. NAS hosts, clients, and other devices can be widely distributed across the entire enterprise network environment. NAS can provide reliable file-level data integration because file locking is handled by the device itself.

NAS is used in efficient file sharing tasks, such as NFS in UNIX and CIFS in Windows NT. Network-based file-level locking provides advanced concurrent access protection.

Iii. DAS

Direct Attached Storage directly attaches Storage, which means that Storage devices are directly connected to a server through bus (SCSI, PCI, IDE, and other interfaces. The cost and configuration of DAS are low, making it attractive for small businesses.

DAS problems:

① The server itself is prone to system bottlenecks.

② If the server fails, the data cannot be accessed.

③ For systems with multiple servers, devices are scattered and cannot be managed. When multiple servers use the DAS, the storage space cannot be dynamically allocated between servers, which may cause a considerable waste of resources.

④ The data backup operation is complex.

Iv. Summary

FC-SAN, IP-SAN, NAS, DAS differences

	DAS	NAS	FC-SAN	IP-SAN
Cost	Low	Relatively low	High	Relatively high
Data transmission speed	Fast	Slow	Extremely fast	Fast
Scalability	Scalability	Relatively low	Easy to expand	Most scalable
Server access storage	Direct access to storage data blocks	Access by file	Direct access to storage data blocks	Direct access to storage data blocks
Server system performance overhead	Low	Relatively low	Low	Relatively high
Security	High	Low	High	Low
Centralized storage management?	No	Yes	Yes	Yes
Backup Efficiency	Low	Relatively low	High	Relatively high
Network Transmission Protocol	None	TCP/IP	Fiber Channel	TCP/IP

:

Differences and relationships between DAS, SAN, NAS, and IPSAN

IP-SAN is actually developed by NAS and SAN. The author can check the development history of NAS and SAN. It is not difficult to see the two teams gradually formed in the early days due to the development of DAS and the development trend of FAS.
NAS network storage, the early NAS does not want to have so many features now. It only provides file-level storage services and supports data transmission in heterogeneous environments. (Heterogeneous environment means that two systems in the network transmit data through NAS. The image point is that if the WINDOWS system wants to access the UNIX system, it must pass through NAS to access the first-hand data. Because the file transfer protocols between the two systems are different)
From a certain point of view, I prefer to transfer storage servers to NAS. He has his own system and storage space.
NAS uses the IP network. He thinks it is a terminal more vividly in the network.
Currently, NAS features are not limited to file-level data storage, but more and more features are added to NAS. (NAS systems are currently linux) So I prefer to define NAS as a file-level server.
The development direction of SAN was not the same as that of NAS in the early days. He prefers fast data block forwarding and secure data transmission.
And the backup data solution is not fully dependent on the server.
SAN is a storage area network. Is composed of optical fiber and SAN Switch SAN card storage. Form a storage network. The role of the server set. He has his own transmission protocol. Unable to work in the outer network. High cost.
The cost for SAN networks is high. All developers are working slowly to reduce SAN network costs. And developed a IP-SAN network.
The IP-SAN is made up of ISCSI card storage. He can work on an IP network. The general workflow is that the terminal transfers SCSI files and encapsulates them into ISCSI format through the ISCSI card. (In an IP network, the SCSI protocol cannot be used. Therefore, we need to package data and use the ISCSI protocol on both sides for data communication)
This reduces the overall cost. Not as dedicated storage networks as SAN does. Fully depends on the IP network. However, the data is backed up. The data volume is very large. All bandwidth that occupies an IP network. Therefore, in the IP-SAN network, the backup link generally needs to re-open a new link rather than completely dedicated IP network resources.
Nas san IP-SAN from the application point of view, it is not difficult to see
SAN is used in large data centers (such as banking and insurance)
IP-SAN is applied to medium-sized enterprises (IP-SAN in a whole large data center cannot be compared with SAN. After all, a dedicated storage area network is much faster than an IP network. But the cost is much lower than that of SAN)
NAS is used in small businesses (databases cannot be backed up. Work only on file-level applications)

What are the differences between ip san, DAS, NAS, and fc san in storage?

Ip san: a storage area network (SAN) using iSCSI technology. The transmission medium is an IP network.
DAS: the most economical structure for servers to directly Mount storage devices.
NAS: a storage device directly attached to the network, which is equivalent to a network file sharing server.
Fc san: it is a SAN network that uses optical fiber technology. The transmission medium is optical fiber, which has the highest performance and is currently the most widely used.