IBM X3650 M3 7945xj9 RAID10 Configuration

IBM X3650 M3 7945xj9 RAID10 Configuration

First, RAID introduction
RAID is an abbreviation for redundent array of inexpensive disks, literally "Redundant array of Inexpensive Disks" or "Disk array". Later, the letter I in the raid was changed to Independent,raid as a "redundant array of independent disks", but this was only a change in the name, and the substance did not change. RAID can be understood as a method of using a disk drive, which links a set of disk drives in some logical way to use as a logical disk drive.
Advantages of RAID
1. High transmission rate. In some RAID modes, you can have many disk drives transfer data simultaneously, and these disk drives are logically a disk drive, so using RAID can reach a single disk drive several times the rate. Because the CPU is growing fast, and the data transfer rate of disk drives cannot be greatly improved, a solution is needed to resolve the contradiction between the two.
2. Higher security. Compared to ordinary disk drives, many RAID modes provide a variety of data repair functions, and when a disk drive in a raid fails to be used, the data in this drive can be recovered through the other disk drives in the raid, and ordinary disk drives cannot be implemented. This is a second reason to use RAID.
Classification of RAID
RAID 0, no redundant, non-verified disk array. The data is distributed across disks with no fault tolerance and the fastest read and write speeds in raid, but because any one disk corruption can invalidate the entire RAID system, the safety factor is less than a single disk. Generally used in the data security requirements are not high, but high speed requirements, such as: large-scale games, graphics and image editing. This RAID mode requires a minimum of 2 disks, while more disks provide more efficient data transfer.

RAID 1, mirrored disk array. Each disk has a mirrored disk, and the mirrored disk remains consistent with the contents of the original disk at all times. RAID1 has the highest security, but only half of the disk space is used to store data. It is mainly used in situations where the data security requirements are high, and the data that is damaged is required to be recovered quickly. This RAID mode requires only 2 disks per group.

RAID 0+1, from its name can be seen, it combines RAID0 and RAID1 technology, the data in addition to distributed across multiple disks, each disk has its physical mirror disk, providing full redundancy capability, allowing one of the following disk failures, without compromising data availability, and have fast read and write capability. However, raid0+1 requires at least 4 disks to be assembled.

RAID 5, parity disk array with no independent check disc. Parity checks are also used to check for errors, but there is no independent check disk, but a special algorithm is used to calculate the location of any stripe check block. This ensures that any read-write operation to the check block is balanced across all the raid disks, which improves system reliability and eliminates bottlenecks, and provides good performance for both reading and writing of large and small amounts of data. In order to be able to write data and check code information across all the disks in the array, the RAID 5 setting requires a minimum of three disks, so in this case, 1/3 of the disk capacity will be used by the backup checksum, and when there are four disks, 1/4 capacity is required as a backup, To minimize the worst-case incidence. When the number of disks increases, the disk capacity consumed by the backup checksum on each disk decreases, but the risk of disk failure increases at the same time, but with two disk failures, data recovery is not possible.

Jbod,jbod (Just bundle of Disks) is a simple disk bundle. JBOD logically concatenates several physical disks one by one, providing a large logical disk. The data on JBOD is simply stored from the first disk, and when the first disk is run out of storage space, the data is stored sequentially from the subsequent disk. JBOD Access performance is exactly the same as access to a single disk, and does not provide data security. It simply provides a way to take advantage of disk space, where JBOD's storage capacity equals the sum of the capacity of all the disks that make up the JBOD.

Matrix RAID, Fabric disk array. is a new Intel-created patented RAID mode for SATA interfaces that is capable of simultaneously implementing RAID 0 and RAID 12 modes on 2 disks, which works by dividing some disk space on each disk in 2 disks into RAID 0 or 1. The remaining space is composed of RAID1 or 0. Matrix raid also features a "hot spare" hard drive that supports RAID 1 array partitioning. Motherboards that typically support the Matrix RAID feature have four SATA interfaces, while a set of matrix RAID requires only two hard drives, using two SATA interfaces. The other two idle SATA interfaces can be plugged into the hard drive, starting the "Hot backup" function. When a hard disk in the matrix RAID system fails, the "Hot spare" hard drive will immediately take over the work to ensure the security of the data in the RAID 1 array partition. Because the data in the RAID 0 array partition is damaged when a hard drive crashes, the "Hot backup" drive is not valid for a RAID 0 array.

Due to the different RAID controller manufacturers and product models, the type of RAID can be supported, only a few motherboards can fully support the above RAID mode, and some other RAID mode is basically used for professional servers and workstations, this is not introduced here.
Disk space usage in RAID mode
The number of disk spaces that can be used in the actual application of different raid modes, as illustrated in the list:

* Matrix raid because of its special disk allocation concept, so here is a separate illustration:
For example, the total capacity of a RAID with 2 120G disks is 240G, a RAID 0 is established, a RAID 0 mode disk capacity is manually specified between 1 and 238G, and then a RAID 1 mode is established using the remaining disk space. Such as:
100G (RAID 0) + 70G (RAID 1) or
50G (RAID 0) + 95G (RAID 1), etc.
How you allocate the disk size for RAID 0 and RAID 1 can be determined by usage requirements.
Second, RAID production
General on-board RAID controller in the motherboard BIOS will have the controller's open and close options, as well as the necessary RAID bios to make the switch options, set them to open and save the BIOS, at post, after the IDE device detection, there will be raid The BIOS self-test interface appears, press the special key to enter the raid BIOS to create, delete, data recovery and so on. Set Block Size (block sizes), generally choose 64K to 128K, large chunks of large files for reading and large games or programs to run good, and small chunks, for small files read or set up a Web, BBS server and other benefits. Then save the raid, and note that the original data on the disk will be erased when the disk is set to raid.

RAID configuration in IBM X3650 M3 server Combat
Http://www.2cto.com/os/201108/101150.html
Background: RAID is a redundant array of inexpensive disks that appears as a separate, large-scale storage device under the operating system. RAID can give full play to the advantages of multiple hard drives, can increase the speed of the hard disk, increase capacity, provide fault-tolerant features to ensure data security, in the case of any one hard disk problems can continue to work, not affected by the damage to the hard drive. RAID can be divided into soft raid and hard raid, the general middle-grade server more use hardware RAID controller, hardware RAID based on RAID card, software RAID based on operating system.
IBM X3650 M3 7945xj9 This model is a batch of special machines, the same as the channel machine 7945i61 configuration, but the price and accessories are much cheaper. M1015 RAID cards support RAID 0, 1, 5, 10. Typically when a hard drive is connected to an array card (RAID), the operating system will not be able to see the physical hard disk directly, so you need to create a RAID so that the system can identify it correctly and then install the system. Prepare RAID10 for this machine as required, as follows:
First, start the machine, the self-test process will have <CTRL>+<H> tips, at the same time press the two keys and then click Start, you can enter the Webbios graphical setting interface. Shown

Second, click on the Configuration Wizard to enter the array setup Guide.

Clear configuration: Clears the existing configuration information, noting that all data is lost.
New configuration: Clears the existing configuration information and creates a new configuration.
Add configuration: Preserves the original configuration information and adds a new hard disk to the original configuration. (This configuration usually does not cause data loss, but the operation is risky and it is recommended to back up the data first!) ）
Note: If you select the top two options (Clear configuration and new configuration), all data will be lost! Please back up all data first! If it's a new machine, don't worry about it. Next, we select New Configuration, click on the left side of the disk and select Addtoarray to add it to the right diskgroups, add two disks to a group, then click Accept DG After creating the first group.

Configuring RAID10 requires the creation of a disk Group with the same number of two physical hard disks as shown in.

Select the appropriate drive and after you have configured disk Group, click the Accept button. Then click on the Next button to go to configure the span interface, select the existing disk Group Click the Add to Span button to join the span. Then click the Next button.

Enter the configuration RAID10 parameter interface and modify the corresponding parameters as required. Parameters such as strip Size and RAID level. (It is generally recommended to select the default setting.) After the configuration is complete, click the Accept button.

Enter the virtual Disk Preview screen and click the Accept button when you are sure.

Prompts to save configuration information. Click the Yes button.

Prompt will lose all data, after confirming, click the Yes button.

The interface is entered after the configuration is complete.

When you click the Home button to return to the main screen, you can see the status of all current drives.

Click Exit to save the exit. Restart the machine, put the system boot disk to install the system normally.
Add:
Several modes of operation for raid
1, RAID0
That is, data stripping is a sub-strip technology. RAID 0 can connect multiple hard drives into a larger drive group, which can improve disk performance and throughput. RAID 0 has no redundancy or bug fixes, is low cost, requires at least two disks, and is generally only used if the data security requirements are not high.
(1), RAID 0 The simplest way
is the X block the same hard disk in the form of hardware through the Intelligent disk controller or the operating system of the disk driver in a software-series together, to form a separate logical drive, the capacity of a separate hard disk x times, when the computer data is written to each disk, when the space of a disk is exhausted, The data is automatically written to the next disk, which has the advantage of increasing the capacity of the disk. The speed is the same as any one of the disks, and if any of these disks fail, the entire system will be compromised and the reliability is the 1/n of a single hard drive.
(2), another way of RAID 0
is to create a stripe set with a reasonable stripe size of n hard drives, preferably with a dedicated disk controller for each drive, which increases n times the speed of reading and writing data to n disks while the computer is reading and writing data. Improve the performance of the system.
2. RAID 1
RAID 1 is called disk mirroring: The data of one disk is mirrored on another disk, and the reliability and repairable of the system can be guaranteed without compromising the performance, it has high data redundancy, but the disk utilization is 50%, so the cost is the highest, and it is used in the case of saving critical important data. RAID 1 has the following features:
(1), RAID 1 each disk has a corresponding mirror disk, at any time the data are synchronized mirroring, the system can be from a set of mirrors on any disk to read data.
(2), the disk can use space only half the total disk capacity, the system cost is high.
(3), as long as there is at least one disk in any pair of mirrors in the system can be used, and even in half the number of hard disk problems when the system can operate normally.
(4), a hard disk failure of the RAID system is no longer reliable, should promptly replace the damaged hard disk, or the remaining mirror disk also has problems, then the entire system will crash.
(5), after the replacement of the original data will take a long time to synchronize the image, the external access to the data will not be affected, but the performance of the entire system has declined.
(6), RAID 1 disk controller load is quite large, with multiple disk controllers can improve the security and availability of data.
3, Raid0+1
Combining RAID0 and RAID1 technology, data is distributed across multiple disks, each with its physical mirror disk, providing full redundancy, allowing one of the following disk failures without compromising data availability and having fast read/write capability. Raid0+1 to establish a stripe set of at least 4 hard disks in disk mirroring.
4, RAID2
The computer saves each bit of data on a disk while writing the data, and the Hamming checksum code obtained by a different bit operation of the data is saved on another set of disks, because it can correct the error in the case of data error, in order to ensure the correctness of the output. However, using data redundancy technology, the rate of output data depends on the slowest disk in the drive group. The RAID2 controller is designed to be simple.
5. RAID3: Parallel transmission with parity check code
RAID 3 uses a dedicated disk to hold all of the checksum data, while creating a read and write operation on the remaining disks for stripe-set distributed data. When the data is read in a good RAID 3 system, it is only necessary to find the corresponding data block in the data storage disk for read operation. However, when writing to raid 3, it is necessary to calculate the checksum of all blocks in the same band as the data block and write the new values back into the check block, which increases the overhead of the system. When a disk fails, all data blocks on the disk must be re-established using the checksum information, and if the data block to be read is located exactly on the damaged disk, all other blocks in the same area must be read together and the lost data rebuilt based on the checksum, slowing the system down. When the damaged disk is replaced, the system must have a block of data to reconstruct the data in the bad disk, and the performance of the whole system will be severely affected. RAID 3 is the biggest disadvantage of the check disk is very easy to become the bottleneck of the entire system, for frequent write operation of the application will lead to the overall performance of the raid system degradation. RAID 3 is suitable for databases and Web servers.
6, RAID4
RAID4 is the independent disk structure with parity code, RAID4 and RAID3 very much like, its access to data is data block, that is, per disk, each time is a disk, RAID4 features and RAID3 also quite like, but in the failure of recovery, it is much more difficult than RAID3 , the design of the controller is much more difficult, and the efficiency of accessing data is not very good.
7, RAID5
RAID 5 Spread the check block across all the data disks. RAID 5 uses a special algorithm to calculate where any of the stripe check blocks are stored. This ensures that any read-write operation to the check block is balanced across all the raid disks, eliminating the potential for bottlenecks. RAID5 's readout efficiency is very high, writing efficiency is general, block-type collective access efficiency is good. RAID 5 improves system reliability, but does not solve the parallelism of data transmission, and the design of the controller is very difficult.
8, RAID6
RAID6 is a separate disk structure with two parity parity codes for distributed storage, which is an extension of RAID5, mainly used to request data absolutely can not error, using two parity values, so need to n+2 disk, while the design of the controller becomes very complex, writing speed is not good, It takes more time to calculate the parity value and verify the correctness of the data, resulting in an unnecessary load that few people use.
9, RAID7
RAID7 is optimized high-speed data transfer disk structure, it all the I/O transfer is synchronous, can be controlled separately, so as to improve the parallelism of the system and the speed of the system Access data, each disk with a buffer memory, the real-time operating system can use any real-time operation of the chip, Meet the needs of different real-time systems. Allows for management and monitoring using the SNMP protocol, you can specify a separate delivery channel for the checksum to increase efficiency. Multiple hosts can be connected, and access time is almost 0 when multiple users access the system. However, if the system loses power, the data in the buffer memory will be lost, so it needs to work with UPS and the RAID7 system is expensive.
10, RAID10
RAID10 high reliability and efficient disk structure it is a band structure plus a mirror image structure, can achieve both efficient and high-speed purposes.


Manually loading the array card driver
V Night God Moon v posted on 2009-10-21 11:25
Http://bbs.51cto.com/thread-627405-1.html
General IBM x Server installation, the new machine will choose to use the Serverguide to install, but if your machine is already in use, but for some reason need to reinstall the system, with serverguide words is not feasible, not to say no, but the original array and other information will be lost , so if manually to load the array card, just need to format the C-disk, and other disk information will be preserved, then see how to do it, according to the following steps: (first remind everyone, manually load the array card, if you are ready to a USB mobile floppy drive or a floppy disk.) ）
First to the official website to download the array card driver, only the English site, you can choose the series. I asked for technical support questions and the following links may be invalid.
IBM ServeRAID M Series and MR10 series SAS Controller driver v4.36 for Microsoft Windows Server 2003/2008-IBM bladecente R and System X
IBM ServeRAID M Series and MR10 series SAS controller drivers v4.36 for Microsoft Windows Server 2003/2008-IBM BladeCenter and System X
http://www-947.ibm.com/support/entry/portal/docdisplay?lndocid=MIGR-5082161
1) Download down, is an EXE file, run it, in fact, is extracted, drawings:

2) Then find the extracted files, find the corresponding model folder, open (note the operating system selection):

3) Then send all the files in the folder of the corresponding array card model to the floppy drive that was connected to it before, or copy the past, of course.

4) OK, just use the installation CD boot up, see this screen to press F6.

5) Wait a minute, you can see the following screen, press "s" to load the additional driver, "ENTER" is not loaded driver, we press "s"

6) After choosing to load the additional driver, the following screen appears, the screen is prompted to insert the previously done array card driver floppy disk, and then press "ENTER" to continue.

7) Then the driver for the array card will appear, press "ENTER" to select the array card.

8) Then the following screen appears, "s" is to select the driver in the floppy disk, "ENTER" is to choose Windows Default driver, of course, press "s" Ah.

9) OK, then you will be prompted if you need to load another boot, and if you do not need to load another driver, press "ENTER" to continue.

10) and then just like the usual system, you can press "ENTER" to continue the installation.

11) Then choose F8 Agree, after the usual no difference, if you load, then will now out of the installation of the partition, such as C.D.E What, will ask you installed on which disk, do not format and so on, if not manually loaded, your previous partition and other information can not be seen.

IBM X3650 M3 7945xj9 RAID10 Configuration