RAID disk array technology and data recovery principle

At present, people have gradually realized the disk array technology. Disk array technology can be divided into several levels of 0-5 RAID technology in detail, and a new level of so-called RAID Level 10, 30, 50 has been developed. RAID is an abbreviation for Redundant Array of Inexpensive Disk. The benefits of using RAID are simply: high security, fast speed, and large data capacity.

Some levels of RAID technology can increase speed to 400% of a single hard drive. Disk arrays work together to connect multiple hard drives, greatly increasing speed and increasing the reliability of hard disk systems to near-error-free realms. These “fault-tolerant” systems are extremely fast and extremely reliable.

From the perspective of the disk array

The most important specification of the disk array is speed, which is the type of CPU. We know that SCSI evolved from SCSI 2 (Narrow, 8 bits, 10MB/s), SCSI 3 (Wide, 16bits, 20MB/s), Ultra Wide (16bits, 40MB/s), Ultra 2 (Ultra Ultra Wide, 80MB) /s), Ultra 3 (Ultra Ultra Ultra Wide, 160MB/s), from SCSI to Serial I/O, also known as Fibre Channel (FC-AL, Fibre Channel - Arbitration Loop, 100 – 200MB/s), SSA (Serial Storage Architecture, 80 – 160 MB/s), when using Ultra Wide SCSI, 40MB/s disk array in the past, the CPU requirements are not too fast, because SCSI itself is not very fast, but when SCSI evolves When it comes to Ultra 2, 80MB/s, the CPU requirements are critical. A general CPU, such as 586, must be changed to a high-speed RISC CPU (such as Intel RISC CPU, i960RD 32bits, i960RN 64 bits), not only RISC CPUs, but also 32bits, 64bit RISC CPUs. The difference between 586 and RISC CPU is imaginable! This is from the perspective of disk array.

From the server's point of view

The structure of the server has been changed from the traditional I /O structure to the I 2 O (Intelligent I /O, I2O for short) structure, the purpose is to reduce the server The burden on the CPU will separate the system's I/O from the server's CPU load. Intel therefore proposed the architecture of I2O, which is also responsible for I/O work by a RISC CPU (i960RD or I960RN). Imagine if the RISC i960 CPU is responsible for I/O in the server. As a result, the 586 CPU is still used on the disk array. Will it be faster?

From the perspective of the operating system

SCO OpenServer 5.0 32 bits

MicroSoft Windows NT 32 bits

SCO Unixware 7.x 64 bits

MicroSoft Windows NT 2000 32 bit 64 bits

SUN Solaris 64 bits ........Other operating systems

The operating system has been switched from 32 bits to 64 bits. The CPU on the disk array must be an Intel i960 RISC CPU to meet the speed requirements. 586 CPU is not enough!

Disk array function

Is the hard disk connection in the disk array connected with SCA-II integral backplane or just SCSI cable? In SCA- II Is there an isolation chip on the overall backplane to prevent the high/low voltage generated by the hard disk during hot plugging, causing the system voltage to reflow, causing system instability and data loss. We must pay attention to this problem, because many hard disks in the disk array share the same SCSI bus

One hard disk is hot swappable, can not affect other hard disks! Is it hot swap or hot swap? The hard disk has a hot-swappable hard disk, the 80-pin hard disk is a hot-swappable hard disk, the 68-pin is not a hot-swappable hard disk, and there is no hot swapping. The difference in design on the circuit is that there is no protection circuit design, the same The hard drive tray is also the same as the difference between true hot swap and fake hot swap.

Is there a sequential requirement for the hard disks in the disk array? That is to say, can the hard disk be inserted back into the array out of order, and the data can still be accessed normally? Many people think that it is not very important, it will not happen. , but it may happen, we have to prevent it from happening. If you use six hard disks for the array, the six hard disks are placed in the disk array in the most initial initialization. The first, second, and sixth hard disks are in order.

If you have a disk array that has sequential requirements, you should pay attention: one day you take out the hard disk, you must insert it into the disk array in the original order when cleaning, otherwise you The data may be due to the hard disk order and the original, the controller on the disk array does not recognize the data loss! Because your hard disk's SCSI ID number is broken. Nowadays, disk array products have such a function that does not require the order of the hard disks. In order to prevent the above events, the hard disks are not required to be in order.

We will discuss these new technologies and the advantages and disadvantages of different levels of RAID. We don't want to cover those critical technical details, but introduce disk arrays and RAID technology to people who are not yet familiar with them. I believe this will help you choose the right RAID technology.

Spanning for Hard Disk Data

Data platter technology enables multiple hard disks to work like a hard disk, which allows users to break through cheaply by combining existing resources or adding resources. Existing hard disk space limitations.

Four 300 megabyte hard drives are connected together to form a SCSI system. The user only sees a C drive with 1200 megabytes instead of C, D, E, F, and four 300 megabyte hard drives. In such an environment, the system administrator does not have to worry about insufficient hard disk security check space on a hard disk. Because the current 1200 megabytes of capacity is all on one volume (such as hard disk C). System administrators can safely build any level of file system they need without having to plan their file system under the constraints of multiple separate hard disk environments.

Hard disk data is not RAID itself, it can not improve the reliability and speed of the hard disk. But it has the advantage that multiple small, inexpensive hard drives can be added to the hard disk subsystem as needed.

Disk Array Classification

Disk Striping (RAID 0)

Hard disk segmentation method writes data to multiple hard disks instead of just one On the disk, this is also called RAID O. In the disk array subsystem, data is written to multiple hard disks in the order of the "segment" specified by the system. For example, data segment 1 is written to hard disk 0, and segment 2 is written. Hard disk 1, segment 3 is written to hard disk 2, and so on. When the data is written to the last hard disk, it resumes writing from the next available segment of disk 0, and the entire process of writing data is repeated until the data is written.

A segment consists of blocks, which in turn consist of bytes. Therefore, when the size of the segment is 4 blocks and the block is composed of 256 bytes, the size of the segment is equal to 1024 bytes according to the byte size. The first to 1024 bytes are written to the disk 0, and the first to the twentieth bytes are written to the disk 1 and the like. If our hard disk subsystem has 5 hard disks, we have to write 20,000 bytes

In short, because the hard disk segmentation method is to write (read) data to multiple hard disks immediately, so its The speed is faster. In fact, the transfer of data is sequential, but multiple read (or write) operations can overlap each other. That is to say, when segment 1 is written to drive 0, the operation of segment 2 writing to drive 1 is also started; and when segment 2 is still writing to disk drive 1, segment 3 data has been sent to drive 2; and so on, At the same time, there are several disks (even if not all) that write data at the same time. Because the data is sent to the disk drive much faster than the physical disk. Therefore, as long as the control software is compiled according to this feature, the above data can be written at the same time.

Unfortunately, RAID 0 is not providing redundant data, which is very dangerous. The calculator can only work properly because the entire hard disk subsystem must be working properly, for example, if a file has segment 1 (in drive 0), segment 2 (in drive 1), segment 3 (in drive 2), then the drive If one of 0, 1, 2 fails, it will cause a problem; if drive 1 fails, we can only get the data of segment 1 and segment 3 physically from the drive. Fortunately, a solution can be found, which is hard disk segmentation and data redundancy.