RAID disk array technology and data recovery principle

At present, the disk array technology is becoming more and more widely known. Disk array technology can be subdivided into several levels of 0-5 raid technology, and has developed the so-called RAID level 10, 30, 50 new levels. RAID is the abbreviation for a low-cost Redundant array of disks (redundant array of inexpensive disk). The advantage of using RAID is simply: High security, fast speed, large data capacity.

Some levels of RAID technology can increase the speed to 400% of a single hard drive. The disk array connects multiple hard drives together, greatly increasing the speed, while improving the reliability of the hard disk system to a near error-Free state. These "fault tolerant" systems are extremely fast and highly reliable.

From the point of view of the disk array

The size of the disk array is most important in the speed, which is the type of CPU. We know that the evolution of SCSI is made up of SCSI 2 (narrow, 8 bits, 10mb/s), SCSI 3 (Wide, 16bits, 20mb/s), Ultra Wide (16bits, 40mb/s), Ultra 2 (Ultra U Ltra Wide, 80mb/s), Ultra 3 (Ultra Ultra Ultra Wide, 160MB/S), in the SCSI to serial I/O, which is called Fibre Channel (Fc-al, Fibre Channel -Arbitration Loop, 100–200mb/s), SSA (serial Storage architecture, 80–160 MB/s), in the past using Ultra Wide SCSI, 40mb/s disk array , the CPU requirements do not have to be too fast, because SCSI itself is not very fast, but when the SCSI evolution to Ultra 2, 80MB/S, the CPU requirements are critical. A generic CPU, (such as 586), must be converted to a high-speed RISC CPU (e.g. Intel RISC CPU, i960rd 32bits, i960rn bits), not only RISC CPUs, but even 32bits, bits RISC CPU differences. 586 and RISC CPU difference is conceivable! This is based on the view of the disk array.

From the point of view of the server

The structure of the server has been changed from the traditional I/O structure to I2O (intelligent I/O, I2O), which is designed to reduce the burden on the server CPU to separate the system I/O from the server CPU load. Intel therefore proposes a I2O architecture, and I2O is also responsible for I/O by a RISC CPU (i960rd or I960RN). Just imagine if the server is already a RISC i960 CPU responsible for I/O, the result of the disk array is still using 586 CPU, faster?

From the point of view of the operating system

SCO OpenServer 5.0 Bits

MicroSoft Windows NT * bits

SCO UnixWare 7.x Bits

MicroSoft Windows NT bit, bits

SUN Solaris-bits ... other operating systems

When the operating system has been transferred from bits to bits, the CPU on the disk array must be an Intel i960 RISC CPU to meet the speed requirements. 586 CPU is unable to meet!

Features of the disk array

Are the hard drives in the disk array connected using the SCA-II overall backplane or are they just connected with a SCSI cable? Whether there is an isolating chip on the SCA-II overall back panel to prevent the high/low voltage generated by the hard disk in hot plug, so that the system voltage backflow, resulting in the system instability, resulting in data loss situation. We must pay attention to this problem because many of the hard drives in the disk array are sharing this same SCSI bus

One hard drive hot plug, can not affect the other hard drive! Hot plug or live plug? hard disk has a hot plug hard drive, 80-pin hard drive is hot plug hard drive, 68-pin is not hot plug hard drive, there is no hot plug, in the circuit design difference is there is no protection circuit design, The same hard drive bracket is also the difference between the real hot Plug and the fake hot plug.

Is there a sequential requirement for the hard disks in the disk array? That is, can the hard disk be inserted back into the array in an orderly manner, and the data is still accessible? Many people think it's not important, it's not going to happen, but it's going to happen, and we have to prevent it. If you use six hard disks to do the array, in the most initialized, this six hard disks are placed in the array in sequence, divided into first, second ... To the sixth hard drive, there is a sequence.

If you buy a disk array that has a sequential requirement, then you should pay attention to: one day you will be hard to remove, do cleaning must be in the original display order into the disk array, otherwise your data may be due to the hard disk sequence and the original Fu, the controller on the disk array is not recognized and data loss! Because the SCSI of your hard disk The ID number is caused by a disorderly fall. Now the disk array products have this does not require the hard disk has the order of functionality, in order to prevent the above events, are not required to have the order of the hard disk.

We will discuss these new technologies and the pros and cons of different levels of raid. Instead of addressing key technical details, we introduce disk arrays and RAID technologies to people who are not yet familiar with them. Believe this will help you choose the right RAID technology.