The concept that was most clear to me this week was the topic of advanced storage. In this lecture, we covered RAID0, RAID1, RAID5, and RAID6. RAID, which stands for Redundant Array of Inexpensive Disks, is a concept of multiple drives being treated as one. This can increase capacity, speed, reliability, or all three together. RAID0 (striping) involves two or more drives and is extremely fast with no redundancy. It focuses on speed and capacity. If one drive fails, the whole thing fails and all the data is lost. This is less reliable and more dangerous than a single disk drive and also does not cost more per GB. RAID1 (mirroring) uses exactly two disk drives, and each block of data is written to both drives. If one drive fails, the system is still running at full speed and no data is lost. With mirroring, the focus is on reliability, and the cost per GB is doubled. With RAID5, three or more drives are required, and the data blocks are spread across all the drives except for one that is used to protect all the data. "Parity" is calculated and stored on the remaining drive and is recorded as "1" if the sum of the data bits is odd and "0" if the sum of the data bits is even. There is a high read performance like striping. With RAID5, if one drive fails, no data is lost and the system is still running but at a reduced speed. There is a reasonable cost per GB and a capacity of (N-1) drives usable for data. RAID6 has a capacity of (N-2) drives usable for data and a performance similar to RAID5. RAID6 is any type of RAID that can continue to perform in the presence of two concurrent disk failures.
RAID is important in the business world because computer storage is applicable to every aspect of business. Any work done on a computer has to do with storage, and it is important to decide how to set up a computer so work is not lost. There are serious storage problems such as needing a large storage capacity, needing high performance, or dealing with risks such as hard disk failure. It is important for businesses to assess whether something is temporary storage, which might mean RAID0 is more applicable since it is only temporary, or whether it matters if the data is lost or not. For example, if it is vital that a system be reliable and there is a real concern about losing, for instance, client data, RAID1 would be most appropriate. Therefore, setting up computer storage to fit the needs of the organization is very important in the business environment.
This website is a RAID tutorial with a description of the different types of RAID. It has a diagram explaining how it works, and below lists the characteristics and advantages of the type of RAID on one side and the disadvantages on the other side. For example, on the first page, there is a diagram explaining RAID0 and below it says that it requires a minimum of two disk drives. On one side, it explains the characteristics and advantages including: simple design, easy implementation, each block of data is writtten to a separate disk drive, no parity calculation, and best performance by spreading I/O load across many channels and drives. It says the disadvantages of RAID0 are that its not fault-tolerant, failure of one drive results in all data being lost, and that it should not be used in mission critical environments. This site is an in-depth look at the topic covered this week in the segment on advanced storage.
Here is a picture showing striping and mirroring:
Works Cited:
http://www.acnc.com/raid.html
http://www.pantherproducts.co.uk/Articles/images/RAID-01.gif

No comments:
Post a Comment