When using the server, we usually use the disk array card, which is a problem for the friend who installed the disk array card for the first time. First we have to choose the mode of the disk array according to our own needs. Disk array modes will now talk about the details:
recent years, in terms of hard disk capacity, access speed or reliability have been greatly improved, but still can not keep up the development of this increase processor Requirements, making the hard drive still a bottleneck in the computer system. In order to solve the application system requirements for high-speed disk access, various measures have been taken. In 1988, Professor D.A. Patterson of the University of California at Berkeley proposed Redundant Array of Inexpensive Disks (RAID). When the host reads data;
RAID ordinary hard drives a disk array, the write data in the host, the host to the RAID controller data to be written into a plurality of blocks of data, and then written to disk in parallel array The RAID controller reads the data scattered on each hard disk in the disk array in parallel, reassembles them and provides them to the host. The use of parallel read and write operations increases the accessibility of the storage system. In addition, RAID disk arrays can also use mirroring, parity, and other measures to improve the system's fault tolerance and ensure data reliability. RAID
according to different methods used, can be divided into six levels 0-5:
RAID 0: When the host requires a data write, the RAID controller data into a plurality of blocks And then write them in parallel to each hard disk in the disk array; when reading data, the RAID controller reads data from each hard disk, restores the data to the original order and transmits it to the host. The advantage of this method is that it uses data blocking and parallel transmission to improve the read and write speed of the host, and the storage space in the disk array is not redundant. However, it does not improve the reliability of the system. When any hard disk media fails, the system cannot recover.
RAID 1: It divides the hard disks in the disk array into the same two groups and mirrors each other. When any disk media fails, you can use the data recovery on the mirror to improve the fault tolerance of the system. . The operation of the data still uses the block-and-parallel transmission method. All RAID
1 not only improves the read and write speed, but also enhances the reliability of the system. However, its disadvantage is that the utilization of the hard disk is low and the redundancy is 50%.
RAID 2 /RAID 4: both levels rarely used in practice, most systems do not support, will not be introduced here.
RAID 3: Like RAID 0, RAID 3 also uses the method of data block parallel transfer, but all the same is that it calculates their parity check after data partitioning, and then divides the block data. Write the parity information together with the parity information to the hard disk array. In this way, the access speed and reliability of the data are improved. When any hard disk in the array is damaged, the original data can be reconstructed by using information on other data disks and parity disks. In terms of disk utilization, RAID
3 is higher than RAID 1, such as an array of 5 hard disks with a redundancy of only 20%. However, RAID 3 also has disadvantages. Since the parity information is fixedly stored on one hard disk, the hard disk is heavily burdened, which creates a new bottleneck.
RAID 5: Similar to the data processing method used by RAID 3, the difference is that it cross-writes the parity information to each hard disk in the array, thus overcoming the bottleneck problem in RAID 3. .
addition to the above, we can also implement software RAID disk array technology. The storage methods provided by the Windows NT operating system, such as disk striping, disk striping with parity, disk mirroring, and duplexing, are actually software implementations of RAID technology. The disk striping corresponds to RAID
0, the disk mirroring and duplexing correspond to RAID 1, and the disk striping with parity corresponds to RAID 5. The biggest advantage of these methods compared to RAID devices is that they are cheap, but the performance is much lower.
in each of the above levels, RAID 5 is the most prominent advantages, important for large-scale application environments can be used as the first choice. However, for most DIY enthusiasts who are looking for speed, a RAID 0 disk array consisting of two IDE hard disks is more suitable.
Now we are HP ML370 G3 server as an example, I used the HP642 RAID card. The installation method is as follows:
First, we will set the boot to CD-ROM boot, and put it into the HP server's own tool disk (including all the drivers, note: different models of server disk arrays are installed differently, some The disk array card needs to copy the RAID card driver in its own CD to the floppy disk and install it. This is just an example. Install the HP642 array card, when you reboot, you will find a self-test to a fast RAID card, then we press F8, this will go to the array card mode setting menu. I only installed two SCSI 36.4G HP original hard drives. The mode I used was RAID1 (because the server I installed had two mirrors of these two hard disks, I don't talk about these modes, they all have them). Select Create Primary Partition and select RAID1 mode. The first step of the hard disk array is completed, then the computer will automatically restart, then there will be no SCSI boot driver, this is not a problem, because we are connected to the RAID card, so SCSI can not find To the hard drive. this is normal. The computer at this time is booted from the CD-ROM drive. HP's boot CD is a fool-like installation. The first step is to choose the system you want to install. I chose WIN2000 Advanced Server Edition. The next step is to choose your partition size. We choose NTFS format, divide the main partition into 8G, and we don't care about the rest. Then the system will prompt you to insert the installation disk of WIN2000 and insert the installation disk. The system will automatically copy the file, and then restart, the following is the installation of WIN2000, this does not need me to talk about it. After installing the WIN2000 system, click Start--Programs---Computer Management---Disk Management to re-partition the remaining space. All the processes are done.