Sunday, July 1, 2012

RAID

Hello everybody. 
can any body tell me about "RAID", or let me know abt. any 
references or tutorial. 
thanx in advance. 


Just do a seach on the web 
About different raid lavel 
There is more info on the web than what you could imagine. And is very well 
Explained.
RAID (Redundant Array of Independent Disks) Primer 


The term RAID was born out of a paper written at the University of 
California Berkeley. At the time disk technology was going in two different 
directions. There were large 14" disk drives with high capacities and large 
physical size(washing machine). There was also an emerging standard of 
lower capacity and physically smaller(5.25" Full Height) disk drives, these 
new smaller drives were also much less expensive per megabyte. The old way 
of storing data on physically large disks was labeled "SLED - Single Large 
Expensive Disk". A new idea of storing data across a number of small disk 
drives to achieve capacity and performance was labeled "RAID - Redundant 
Array of Inexpensive Disks" Note the title uses Independent not Inexpensive 
, seems the marketing people got invloved in what would be politcally 
correct. The concept behind the Berkeley paper was that high performance, 
capacity and/or redundancy could be acomplished through any of six 
different RAID levels. The levels defined were; RAID 0, RAID1, RAID2, 
RAID3, RAID4 and RAID5. These RAID levels are in fact what is used today. 
RAID 2 was not technically feasible, RAID 0+1 was born later and a 
multitude of marketing RAID levels (RAID 7, RAID 5 +1 and so on) have also 
hit the streets. The bottom line is that the core understanding of the RAID 
levels rests with RAID 0 through 5. It should also be noted that the 
University of California Berkeley never actually created RAID sets of any 
kind, their work waited as a paper tiger until some years later the 
computer industry found a way to make money with those ideas. 


This paper will explain the base function and data layout for the different 
levels of RAID. Each level or type of RAID data layout has specific 
performance and redundancy characteristics. These characteristics allow you 
to custom configure RAID sets to your performance criteria. Understanding 
the basics of RAID data layout is the first step to purchasing a RAID 
subsystem that meets today's and tomorrow's storage requirements. A second 
paper, titled "Buying RAID" goes into more specific hardware configurations 
and features. We invite you to download "Buying RAID" after reading this 
piece. 


Let's get some terminology out of the way. 


ACCESS TIME The time it takes to find, recover and start delivering data. 
If we were talking about a single disk, "access time" would be the time the 
disk took to receive the command, move the heads to the correct position, 
wait for the disk to rotate to the correct location and start to read. 
Access time is particularly important to multi-user and database 
environments. These environments have many many requests to the same data. 
These requests do not necessarily involve long data transfers. "Access 
Time" is important when deciding what level of RAID you will implement. 


ARRAY An "array" is a grouping of disk drives. There could be two disk 
drives in the "array" or there could be 100 disk drives in the array. 


CHUNK The breaking up of data so that it can be stored across multiple disk 
drives. Imagine breaking up a 80 Kb file into five 16 Kb pieces. These 16 
Kb pieces would be referred to as "chunks". Another commonly used term is 
STRIPE. 


COST PER MEGABYTE PENALTY Term used to indicated what the percentage of 
total storage is dedicated to providing data redundancy. 


DISK SET A "disk set" is a specific number of drives grouped together with 
a single characteristic (e.g., RAID 0 , RAID 5). A "disk set" can encompass 
a whole "array" or be a subset of the "array". In other words, there can be 
multiple "disk sets" in an array. A "disk set" will present itself to an 
operating systems as an individual disk drive. This will be important later 
when you are actually buying and configuring a RAID subsystem. "Disk Set" 
is also referred to as RAID SET. 


DUPLEXING "Duplexing" is when members of a "disk set" are spread across 
different SCSI busses. This is important for two reasons, it relieves 
dependance on one SCSI bus in the event of failure and increases 
performance by moving data across two different busses simultaneously. 


MIRRORING Data is stored twice (or more) on two or more different disk 
drives. Shadowing is a commonly shared term with "mirroring". Easy 
definition - What is written on one disk is written on another disk. 


PARITY Parity is a mathematical equation that allows data to be checked for 
integrity. It should be noted that the term parity is loosely used, the net 
effect is that data is generated that allows the stored data to be checked 
for integrity. 


PARITY DISK RAID levels 3, 4 and 5 have a dedicated drive for storing 
parity information. This information may stay on one disk or move between 
disks as parity chunks. 


PARTITION The breaking up of a "disk set" into smaller segments. The 
smaller segments will appear as individual disk drives to the host, while 
still maintaining the RAID properties of the "disk set". 


RAID SET A "RAID SET" is a specific number of drives grouped together with 
a single characteristic (e.g., RAID 0 , RAID 5). A "RAID Set" can encompass 
a whole "array" or be a subset of the "array". Another words there can be 
multiple "disk sets" in an array. A "RAID Set" will present itself to an 
operating systems as an individual disk drive. This will be important later 
when you are actually buying and configuring a RAID subsystem. "RAID Set" 
is also referred to as DISK SET. 


STRIPING The breaking up of data so that it can be stored across multiple 
disk drives. Imagine breaking up a 80 Kb file into five 16 Kb pieces. These 
16 Kb pieces would be referred to as "stripes". Another commonly used term 
is CHUNKS. 


TRANSFER RATE How fast the RAID set or subsystem can transfer the data to 
the host. "Transfer Rate" is important when large contiguous blocks of data 
are being used. Video and image files are examples of large contiguous file 
transfers that occur in streaming mode. "Transfer Rate" is important when 
deciding what level of RAID you will implement. 

RAID 0 =A0- =A0Data Striping 


RAID 0 allows a number of disk drives to be combined and presented as one 
large disk. RAID 0 does not provide any data redundancy - if one drive 
fails, all data is lost. 


Access Time Very Good 
Transfer Rate Good 
Redundancy None 
Cost Per Megabyte None 
Penalty 
Applications Large disk requirements, high performance databases 


RAID 1=A0- =A0Disk Mirroring/Disk Duplexing 


RAID 1 mirrors (shadows) one disk drive to another. All data is stored 
twice on two or more identical disk drives. When one disk drive fails, all 
data is immediately available on the other without any impact on the data 
integrity - performance in degraded mode is also degraded. Performance is 
gained by splitting of functions. If multiple read requests are pending, 
the RAID controller will allows reads from different disk drives. If one 
disk is busy writing the other disk drive can supply read data, at a later 
time. The RAID controller will update the read drive with data from the 
already written disk drive. If each disk drive is connected with a separate 
SCSI channel, this is called "Disk Duplexing" (additional security and 
performance). RAID 1 represents a simple and highly efficient solution for 
data security and system availability. Use RAID 1 when large volumes of 
data are not required. 


Access Time Very Good 
Transfer Rate Good 
Redundancy Yes 
Cost Per Megabyte 100% or more 
Penalty 
Applications Small disk capacities that require redundancy 


RAID 0 + 1 =A0- =A0Combination of RAID 1 and RAID 0 


The idea behind RAID 0+1 is simply based on the combination of RAID 0 
(Performance) and RAID 1 (Data Security). RAID 0+1 disk sets offer good 
performance and data security. Similar as in RAID 0, optimum performance is 
achieved in highly sequential load situations. The major draw back is a 
100% "Cost Per Megabyte Penalty". 


Access Time Very Good 
Transfer Rate Good 
Redundancy Yes 
Cost Per Megabyte 100% 
Penalty 
Applications Multiuser environments, database servers, file serving, web 
site hosting 



RAID 3 Data Bit Striping With a Dedicated Parity Drive 


The data is striped at a byte/bit level across the disk drives. 
Additionally, the controller calculates parity information which is stored 
on a separate disk drive (aP, bP, ...). Even when one disk drive fails, all 
data is fully available. The missing data can be recalculated from the data 
still available and the parity information. This data calculation can also 
be used to restore data to a replaced defective disk. Because the data must 
be presented at the same time, the disk drive spindles must be synchronized 
for RAID 3 to be effective. This represents a practical implementation 
problem for RAID 3. Many RAID controller manufacturers are moving to a RAID 
4 solution or using the term RAID 3 merely as a recognized marketing term 
for high data transfer capability. 


Access Time Good 
Transfer Rate Very Good 
Redundancy Yes 
Cost Per Megabyte Varies. 5 drive set =3D 20%, 6 drive set =3D 17%, 10 
drive set =3D 10% 
Penalty 
Applications Imaging, geological, seismological, video 





RAID 4=A0- Data Striping With a Dedicated Parity Drive 


RAID 4 works just like RAID 0. The data is striped across disk drives. 
Additionally, the controller calculates parity information which is stored 
on a separate disk drive (P1, P2, ...). Even when one disk drive fails, all 
data is fully available. The missing data can be recalculated from the data 
still available and the parity information. This data calculation can also 
be used to restore data to replaced defective disk. RAID 4 offers excellent 
transfer rates when used with large contiguous blocks of data. When used 
with with many small data blocks, the parity disk drive becomes a 
throughput bottle-neck because of it's fixed position.A RAID 4 disk set can 
only lose one disk from it's RAID set. Loosing another disk drive, before a 
replacement is restored, will loose all data in the RAID set. 


Access Time Good 
Transfer Rate Very Good 
Redundancy Yes 
Cost Per Megabyte Varies. 5 drive set =3D 20%, 6 drive set =3D 17%, 10 
drive set =3D 10% 
Penalty 
Applications Imaging, geological, seismological, video 




RAID 5=A0- =A0Data Striping with Striped Parity 


The data is striped across disk drives. Unlike RAID 4, the parity data in a 
RAID 5 set is striped across all disk drives. RAID 5 is designed to handle 
small data blocks. This makes RAID 5 the level of choice for multitasking, 
multiuser and database environments. RAID 5 offers the same level of 
security as RAID 4:=A0when one disk drive fails, all data is fully available, 
the missing data is recalculated from the data still available and the 
parity information. This data calculation can also be used to restore data 
to replaced defective disks. RAID 5 is particularily suited for systems 
with medium to large capacity requirements, with their "Cost Per Megabyte 
Penalty" is relatively low. A RAID 5 disk set can only lose one disk from 
it's RAID set. Loosing another disk drive, before a replacement is 
restored, will loose all data in the RAID set. 


Access Time Very Good 
Transfer Rate Good 
Redundancy Yes 
Cost Per Megabyte Varies. 5 drive set =3D 20%, 6 drive set =3D 17%, 10 
drive set =3D 10% 
Penalty 
Applications Multiuser environments, database servers, file 
serving, web site hosting 

  I was trying to go throurgh SVM. I want to know how we can mirror the entire disk (is it by making the same partition on another disk and making the slice by slice metadeives and then mirroring ) Also for RAID 5 how can we include 3 different disk (not only slice) into a RAID 5 and how to remove the faulty disk and add the new disk to it. Many Thanks for your help and time Pradeep P 



JBOD=A0- =A0Just a Bunch Of Disks 


An allowance was made by virtually all RAID control manufacturers for 
adding a single disk inder the RAID controller that would not be a part of 
any RAID Set. A "JBOD" disk drive appears to the host as an add-on disk 
drive. Using JBODs is a convenient way of adding quick storage. If a JBOD 
disk drive breaks all data is lost 





Access Time Good 
Transfer Rate Good 
Redundancy No 
Cost Per Megabyte None 
Penalty 
Applications Quick increase in capacity 


Mixing RAID Sets in a Disk Array 


The following diagram illustrates that within a Disk Array many different 
RAID levels and capacities can be configured. This allows a RAID user to 
custom configure RAID level characteristics with the user's many different 
performance and capacity requirements. 


We hope this RAID Primer has helped take some of the mystery out of exactly 
what is a RAID level. A following white paper, "Buying RAID" discusses 
architecture, implementations and features of RAID subsystems. The "Buying 
RAID" will help checklist capacity, growth, clusterability, caching along 
with other shopping tips. 


=A0thkz. 

0 comments:

Post a Comment

 
Design by BABU | Dedicated to grandfather | welcome to BABU-UNIX-FORUM