1,RAID 0RAID 0, most commonly known as striping, provides improved I/O rates (in terms of IOPS) by striping the data across multiple drives, allowing the read and write operations to be shared amongst the drives inside the array.This level of RAID provides the best performance for both read and write operations, butprovides no redundancy or protection against data loss. In the event of a single disk failure in the array, all of the data is lost.
one task A,have a+b, for raid0:risk 1:a,risk 2:b ,so:
n disks: n(read+write)
2,RAID 1, most commonly known as mirroring, provides protection against the loss of
data from a single disk by mirroring the writes to a second disk but doesn't provide added write performanceto the system, since the maximum throughput for the array is limited to the I/O capacity
of a single disk
one task A,have a+b, for raid1: write: risk 1:a+b,risk 2:a+b ,so:
n disks: (n read + n/2 write)
3,RAID 10
RAID 10, or RAID 1+0, is a nested RAID level known as a "striped pair of mirrors." It
provides redundancy by first mirroring each disk, using RAID 1, and then striping those
mirrored disks, with RAID 0, to improve performance.
one task A,have a+b, for raid10: write: risk 1:a,risk11:a, risk 2:b ,risk:22 b
so:
n disks: (n read + n/2 write)
4,RAID 01
RAID 01 or RAID 0+1, is a nested RAID level also known as "mirrored pairs of striped
disks." In a RAID 0+1 configuration, the nested levels are the opposite of the RAID 1+0,
with the disks first being striped in a RAID 0 and then mirrored using RAID 1. However,
this type of configuration only allows for a single disk loss from one side of the array,
since a single disk failure in a RAID 0 array causes that array to fail. A loss of a single disk
in both of the RAID 0 striped arrays would result in total data loss.
5,RAID 5 (and 6)
RAID 5 is commonly known as "striping with parity;" the data is striped across multiples
disks, as per RAID 0, but parity data is stored in order to provide protection from single
disk failure. The minimum number of disks required for a RAID 5 array is three. The
parity blocks containing the parity data are staggered across the stripes inside the array
n disks: ((n-1) (read +write)
Raid 0 –每个磁盘的I/O计算= (读+写) /磁盘个数
Raid 1 –每个磁盘的I/O计算= [读+(2*写)]/2
Raid 5 –每个磁盘的I/O计算= [读+(4*写)]/磁盘个数
Raid 10 –每个磁盘的I/O计算= [读+(2*写)]/磁盘个数
由上知，raid5与raid10比较，raid5对单块磁盘的I/O要求较高，常采用raid10
虽然raid5奇偶性是实现容错的一种不错的方式。但是从磁盘写入来说代价高昂。也就是说对于每一个IOP写请求，RAID5需要4个IOPS。
为什么需要这么高写入代价的过程如下:
- 读取原始数据(1 iop)
- 读取当前奇偶数据(1 iop)
- 比较当前数据和新写入请求
- 基于数据差异计算新的奇偶值
- 写入新数据(1 iop)
- 写入新的奇偶值(1 iop)
6,choose what?
Data files:
1,depends heavily on the read-to-write ratio for the .SQL Server tracks the I/O usage of the database files for an instance and makes this information available in the sys.dm_io_virtual_
file_stats Dynamic Management Function. While this information can be used
to determine if your overall workload is heavier for reads vs. writes.
For a database that is primarily read-only, RAID 5 or RAID 6 can offer good read performance.
2,RAID 5 or 6 arrays are commonly used for data warehouses, or for storing data where write latency doesn't impact overall system performance.
3,For OLTP implementations of heavy-write databases, RAID 1+0 provides the best performance
Log files:
Since the transaction log is written to sequentially, RAID 1 can be used in most situations.the transaction log for each database should be located on dedicated physical disks
tempdb:As a general rule, the tempdb database files should be physically separate from the user
data files and transaction log files, on a dedicated disk array. Since tempdb is a writeheavy
database, RAID 1 or RAID 1+0 are usually the configurations best able to support
the concurrent workload of tempdb.since tempdb is used for temporary storage only, Solid State Disks and even RamDisks can be used to significantly improve the I/O characteristics of the tempdb database
http://blog.tianya.cn/blogger/post_show.asp?idWriter=0&Key=0&BlogID=854352&PostID=20759437