Jump to content

Question about RAID !!


Recommended Posts

hi guys question about RAID !!

i just bought a raid card the supports this

Supports RAID 0, RAID 1, RAID 0+1 (Optional)

i am wondering if i raid 2 drives RAID 0 mirror and install windows on it and work on that if later down the line i want to add an extra one on RAID 0 do i need to reinstall everything or can i just plug it in and it will start distributing data over the new HD as well

this is the specification of the PCi card i bought


This Serial ATA host controller card is designed to offer a high performance,

cost effective and reliability solution to user who needs to accommodate storage peripherals with the new Serial ATA interface,

It can control two independent Serial ATA channels, each channel has its own Serial ATA bus and will support one Serial ATA device.

This card supports the Serial ATA (Generation 1) transfer rate of 1.5Gb/s (150MB/s).

Made in China


Silicon Image Sil3114 SATA controller chip

Compliant with PCI Specification, Rev. 2.3, 32bit, 33/66MHz

Compliant with Serial ATA 1.0 specification

Supports RAID 0, RAID 1, RAID 0+1 (Optional)

HDDs Function Normally When Not in RAID Sets

Automatically Select Highest Available Transfer Speed for All ATA and ATAPI Device

Supports SATA up to 150MB/sec

Supports all UDMA and PIO Modes

Supports up to 4 SATA Device

Supports ACPI and ATA/ATAPI

Supports Windows 98, 98SE, Me, 2000, XP, 2003 and NT4.0, Linux and NetWare

RAID Functions(Optional):

Supports RAID functions 0 or 1

Supports co-exist RAID set and Non-RAID HDD

Allow HDDs with different sizes to be configured in RAID set

Package contents:

User Manual

Serial ATA host controller card

4x Serial ATA cable

4x Molex to SATA power cable

Driver CD

Link to comment
Share on other sites

From what I have read.. if you add an extra drive you are going to have to reinstall..

Here are some things to think about.. Raid 0 no back up! one drive fails everything goes.  So unless you have server class hard drives I wouldnt even try it on a consumer drive.  The drive would most likely die with in a year or two. due to the extra writing and strain.

Raid 1 would be better in my opinion you are going to get fast read speeds.. which is what you would be doing more anyways.. and you would not have to worry about a drive craping out on you.

Link to comment
Share on other sites


Here we go..

RAID 1, Disk Mirroring

Up next is the most simple RAID config of them all, RAID 1, aka Disk Mirroring.  The title tells all: you have two or more disks, and you mirror the contents of one onto the other.  With 100% data redundancy, little to no restoration of data on a single disk failure is necessary ... simply copy the pertinent files back from its mirrored twin to the replacement spindle. Or, when you're running an OS or special RAID controller that is mirror aware, simply "break" the mirror and you're set--no more downtime than it takes to reboot or reconfig your controller while another hard drive makes its way to you.

You can already see that there's no speed benefit here on account of the absence of striping.  Rather, due to Error Checking/Correction (ECC), write performance can often be slower than what you'd get with just one simple hard drive not being mirrored at all.  Much of this inefficiency is related to the fact that the mirroring work is  often placed on the CPU(s).  Whether it be in software, or cheap RAID hardware,  if your mirroring isn't handled by the controller, you may actually loose performance. NT/2000 users should be wary on account of the ease with which one can implement RAID 1 in software--it'll slow ya down, for sure.

Now here's an oddity: a read transaction can theoretically occur twice as fast as on single disk. Hence RAID 1 is often used on low-end web servers. The read performance is standard, if not better than single disk performance, and the poorer write performance is largely irrelevant on most web servers (save those doing transactions, of course). RAID 1 configs are great for mid-volume FTP servers as well.

In a nutshell, folks turn to RAID 1 when data integrity is at much more of a premium than I/O performance.  RAID 1 is relatively cheap and quite easy to utilize, and for the cost of a decent backup tape solution, another disk for RAID 1 might be right up your alley.

I should note that this discussion is based on the more recent, er, modern definition of RAID 1.  The original model for this config actually included striping (as in RAID 0), and not simply "disk duplexing." In the end, however, the duplexing model is what the industry uses, and RAID 1 is synonymous with that. Therefore, notice that RAID 1's contribution to the world of storage technology is the principle of data mirroring. With striping and mirroring covered, there's actually only one more basic technological concept to cover: parity.


RAID 1 uses a technology called mirroring or disk shadowing. RAID 1 requires a minimum of two drives that are exactly the same size. Every time a write is executed the same data is written to both drives, i.e. a mirror image. Well, almost a mirror image. The data is not reversed in the same way as when you look in the mirror.

So what you achieve with RAID 1 is data redundancy. If one of the drives fails, the system can continue to run by just writing to one drive. If you have hot swappable drives, you could pull out the bad drive, plug in a new one and the system is back to its normal state. How efficient and easy it is to execute all of this depends on the RAID controller and/or software that is being used.

The disadvantage of RAID 1 is that you lose half of your disk capacity. If you have two 4 GB drives, you don't have a total of 8 GB of space, but only 4 GB. So you are losing half of the capacity of disk space that you paid for. But on the other hand disk drives are fairly inexpensive today. What has to be considered, is what is the cost of downtime if a drive fails on your server. The downtime cost is probably much more than the cost of the additional drive.

Another slight disadvantage is that writing data will result in a slight decrease in performance, as the same data is written twice. But the offset for this is that the reading of data will realize an increase in speed. Basically the slower writing and faster reading offset each other.


Link to comment
Share on other sites

  • Create New...