RAID – The love-hate relationship, but for how long will it last?

Posted by Karim Buzdar on January 25, 2018
Share on Facebook0Share on Google+0Share on LinkedIn0Share on Reddit0Tweet about this on Twitter0
2.75/5 (4)
2.75/54

Back in the day, RAID (Redundant Array of Independent Disks) was considered the king of the storage landscape. Businesses paid a lot to receive the much expensive privilege of data protection. In today’s world though, the once-hot breakthrough technology is considered a sin among most storage circles.

Commonly used as a secondary data protection technology, it helps increase the performance and reliability of data storage. On a physical level, it’s a logical configuration of multiple physical drives to provide fault tolerance and data optimization in a system. One can organize data in a RAID array in different levels by balancing data and recovery needs or speed and fault tolerance properties – essentially whatever factors are most critical to the business.

 

Advantages and disadvantages of various RAID levels

Here’s a list of the most used RAID levels along with respective advantages and disadvantages:

RAID Level 0:

RAID 0, often referred to as striping, is known for higher data throughput as it splits data across any number of disks resulting in access to enhanced speed and capacity. However, at this level, no parity information is stored, leading to no support for fault tolerance or redundant data. Since all discs appear as a set if one fails it results in a broken array and data loss. One can safely say, when it comes to reliability, RAID 0 is a no-go.

RAID Level 1:

This level is also known as mirroring; where data is written twice on the data drive as well as a mirror drive to ensure redundancy. This also helps for speedy data recovery. Since storage capacity becomes only half of the drive capacity, evidently, this level should not be used if storage capacity is a preference. Another downside to this level is that hot swaps aren’t always allowed in case of drive failure.

RAID Level 5:

Most businesses prefer to use this raid configuration as it provides both performance and fault tolerance. The level writes both the data and parity data onto 3 or more discs hence the name striping with parity. The parity data is stored to ensure recovery and fault tolerance. In case a disc fails, the data is recovered seamlessly and the faulty disc can be swapped easily without disturbing operations. The only hit one receives is the lag in performance due to major write operations owing to the parity overhead.

RAID Level 6:

RAID level 6 is a more secure and reliable version of RAID level 5 as it has an additional parity block – meaning if 2 discs fail, the array will remain unbroken. RAID 6 is therefore also known as striping with double parity and ensures redundancy and read performance. However, the performance takes a huge it owing to write operations due to double parity.

RAID Level 10:

The mirroring of RAID 1 combined with the striping of RAID 0 forms RAID 10. It combines the best of both worlds by pairing up the redundancy value from RAID 1 and high performance of RAID 0. So if you need speed as well as reliability, RAID level 10 is the right choice for you. The downside is low usable capacity owing to mirroring resulting in increased cost along with limited options for scalability.

Is RAID right for your business?

Today’s customers are used to accessing data instantly without any interruptions or errors. Unlike most of the recent storage technologies, RAID can never be used to store mission-centric and critical business files.  Reason being, you may have protection against hardware failure but when it comes to protection against corrupt files, errors or malicious activity, you have serious vulnerabilities.

Needless to say, that as disk capacity has continued to increase over the decades, the disc fix time has also increased. Where capacity has shown a 100-fold increase starting from the 1.26GB IBM 3380 to the range of 146GB Savvio 15K drives, the performance increase has only moved up to 50-folds.

Since there is an increased stress on the array when data bits are gathered to rebuild the failed and erroneous disc, there is the potential risk of double-disk fault and read error – hence the shift from RAID to more recent data storage mechanisms.

 

Conclusion

Some say it may take some time to say goodbye to RAID but there are some evident and growing challenges around common RAID levels that are pushing storage makers to turn to alternative data protection mechanisms and rightfully too.

Will the new and comparatively expensive alternatives like erasure coding, local reconstruction codes and Redundant Area of Independent Nodes prove to be more reliable and can RAID keep up with the latest generation of high capacity hard drives? – We think not. The new storage technology may be expensive but it definitely serves us with better data protection and high performance. So who will be the new king of the storage landscape; only time will tell.

Views All Time
5
Views Today
14

Please rate this

To download the software products, please, make your choice below. An installer link and a license key will be sent to the e-mail address you’ve specified. If you consider StarWind Virtual SAN but are uncertain of the version, please check the following document Free vs. Paid. The recent build of Release Notes. A totally unrestricted NFR (Not For Resale) version of StarWind Virtual SAN is available for certain use cases. Learn more details here.



Return to all posts

How to configure a Multi-Resilient Volume on Windows Server 2016 using Storage Spaces
The following two tabs change content below.
Karim Buzdar
Karim Buzdar
Karim Buzdar holds a degree in telecommunication engineering and is a Microsoft Certified Solutions Associate (MCSA) for Server Infrastructure. As an IT engineer and technical author, he focuses on Microsoft Directory Services and PowerShell