Explain different technical ways of storage network technology
2018-05-24 04:01:30
I often receive letters from users asking about the differences between Storage Area Network (SAN), Network Attached Storage (NAS), and Fixed Content Addressing (CAS) technologies that have emerged in recent years. Do not underestimate these problems, if you really understand the essential differences between these three types of technology, you can probably touch the future development of the storage network "pulse". SAN storage system is the most widely used network storage solution. The annual sales volume of SAN hardware equipment exceeds more than 10 billion US dollars of NAS products. This kind of sales performance is very surprising, because, compared with NAS, SAN has many fatal weaknesses, for example, the price is more expensive, the system is more complex, the intelligence is lower, and the SAN system is not Suitable for processing application data. Of course, the success of the SAN is not entirely unreasonable. Its greatest advantage is that it can be compatible with a variety of application software that optimizes performance. In addition, NAS devices have only become popular in the past two years. SAN technology has been born for a long time and has been baptized with time. It is a fairly mature technology and is being gradually applied to business-critical application storage systems. .
On the other hand, comparing the NAS storage device with the SAN, the former has a higher intelligent program, and it is more accurate in terms of file and file attributes. The fundamental difference between a SAN and a SAN is the method of accessing data. If you are accessing data stored on a SAN system, users must go through an application server to obtain file or database information; if you are accessing data on a NAS system, users can directly call data on the network without using the application. Program server. Therefore, when the user needs to share files with others frequently at work, it is more convenient to use the NAS system. In addition, NAS supports higher-level NFS files and CIFS file access protocols, and the response time for reading and writing data is relatively shortened.
In addition to NAS and SAN, there is another mainstream storage system - Fixed Content Addressing (CAS). EMC's Centera series of disk arrays, introduced several years ago, is the industry's first object-based, fixed content addressing system that talks to a variety of applications through an exclusive and patented application programming interface (API). In other words, when third-party software vendors design products, they must integrate Centera's API standards in order to be able to take advantage of record retention strategies. Alternatively, users can also use the Centera Universal Access (CUA for short) device, which supports the standard NFS and CIFS file protocols and allows non-integrated applications to use Centera. The so-called fixed content addressing means that the system assigns a unique address for each saved object and its metadata. No matter how many times the object is used, only one original is saved and no duplication occurs. All objects and metadata are archived and retrieved in a large, flat address space. The biggest reason why the fixed content addressing system is widely used is that it allows the user to set certain data management policies and prohibits any person from arbitrarily changing or deleting files within a specified period of time, thereby helping the user to deal with related issues. The purpose of laws and regulations.
While fixed content addressing (CAS) is an emerging storage technology, it is not new. Storage systems based on NAS technology have most of the functions that CAS has. The high-end CAS system has very impressive scalability, which is unmatched by traditional NAS products. However, who can guarantee that the next generation of NAS products will not surpass it? In fact, most of the competing products that currently “collide†with Centera are based on NAS technology. They have a scalable file system and a clustered network architecture. Since the introduction of Centera, sales have been skyrocketing. The reason why it has achieved such excellent results is that, in addition to its unique system architecture and high expansion performance, a large number of software tools are integrated on the platform, which is also one of the important reasons. At present, the number of software that can support the EMC CenteraAPI standard has exceeded 200 kinds, and this number is still growing.
In addition, IBM and SGI and other industry tycoons have also launched their own SAN file systems, making the market landscape more complex. These products are actually just a kind of network file system that can talk in Fibre Channel. It has the file-level intelligence of NAS system, but it belongs to the genuine SAN system components.
Through the above introduction, I believe readers have all realized that files, file attributes, and metadata are the key factors that affect the management, protection, and disaster recovery of various stages of the data life cycle.
On the other hand, comparing the NAS storage device with the SAN, the former has a higher intelligent program, and it is more accurate in terms of file and file attributes. The fundamental difference between a SAN and a SAN is the method of accessing data. If you are accessing data stored on a SAN system, users must go through an application server to obtain file or database information; if you are accessing data on a NAS system, users can directly call data on the network without using the application. Program server. Therefore, when the user needs to share files with others frequently at work, it is more convenient to use the NAS system. In addition, NAS supports higher-level NFS files and CIFS file access protocols, and the response time for reading and writing data is relatively shortened.
In addition to NAS and SAN, there is another mainstream storage system - Fixed Content Addressing (CAS). EMC's Centera series of disk arrays, introduced several years ago, is the industry's first object-based, fixed content addressing system that talks to a variety of applications through an exclusive and patented application programming interface (API). In other words, when third-party software vendors design products, they must integrate Centera's API standards in order to be able to take advantage of record retention strategies. Alternatively, users can also use the Centera Universal Access (CUA for short) device, which supports the standard NFS and CIFS file protocols and allows non-integrated applications to use Centera. The so-called fixed content addressing means that the system assigns a unique address for each saved object and its metadata. No matter how many times the object is used, only one original is saved and no duplication occurs. All objects and metadata are archived and retrieved in a large, flat address space. The biggest reason why the fixed content addressing system is widely used is that it allows the user to set certain data management policies and prohibits any person from arbitrarily changing or deleting files within a specified period of time, thereby helping the user to deal with related issues. The purpose of laws and regulations.
While fixed content addressing (CAS) is an emerging storage technology, it is not new. Storage systems based on NAS technology have most of the functions that CAS has. The high-end CAS system has very impressive scalability, which is unmatched by traditional NAS products. However, who can guarantee that the next generation of NAS products will not surpass it? In fact, most of the competing products that currently “collide†with Centera are based on NAS technology. They have a scalable file system and a clustered network architecture. Since the introduction of Centera, sales have been skyrocketing. The reason why it has achieved such excellent results is that, in addition to its unique system architecture and high expansion performance, a large number of software tools are integrated on the platform, which is also one of the important reasons. At present, the number of software that can support the EMC CenteraAPI standard has exceeded 200 kinds, and this number is still growing.
In addition, IBM and SGI and other industry tycoons have also launched their own SAN file systems, making the market landscape more complex. These products are actually just a kind of network file system that can talk in Fibre Channel. It has the file-level intelligence of NAS system, but it belongs to the genuine SAN system components.
Through the above introduction, I believe readers have all realized that files, file attributes, and metadata are the key factors that affect the management, protection, and disaster recovery of various stages of the data life cycle.