This post is based on an article titled Alternatives to RAID, written by Marc Staimer, which appeared in June 2010 issue of Storage magazine.
RAID (Redundant Array of Independent disks) have been around for more than 20 years to protect data in intelligent storage systems. In case of disk failure, disk data is recreated using parity disk or from its mirror. As we add more hard disk in the storage array, the MTBF (Mean Time Between Failure) decreases. With the increase in hard disk size, the rebuild also takes longer. There are tradeoffs in each RAID type in terms of write performance, read performance, level of data protection, speed of data rebuilds and the usable storage. RAID 1, 10 provide guaranteed data availability, but at a severe costs. In addition to that, system resources are required for every copy. In RAID 5, the data will be rebuilt on a spare drive. In this process, every sector on every hard disk drive will be used, resulting heavy utilization of the other HDDs in the RAID set and likelihood of another HDD failure. RAID 6 has two parity to protect data even if two hard disks fail at the same time, but this consumes additional usable capacity. The system performance takes a big hit if two disk drives need to be reconstructed.
Due to the above mentioned issues, Other alternatives are becoming available, two of them are:
RAID + Innovation:
IBM’s EVENODD and NetApp’s RAID-DP have enhanced RAID 6 by reducing algorithm overhead while increasing performance. NEC Corp.’s RAID-TM or triple mirror aims to solve RAID 1 data loss risk if both the primary and mirror drive fail or if there’s a non-recoverable read error. RAID-TM writes data simultaneously to three separate HDDs so if two HDDs fail or there are unrecoverable read errors in the same mirror, the app still has access to its data with no degradation in performance even as the drives are rebuilt. The advantage is performance; the disadvantage is far less usable capacity.
RAID-X is an IBM XIV Storage System innovation that uses a wide stripe to reduce RAID tradeoffs of performance and data loss risk. It is a variation of RAID 10 and uses intelligent risk algorithms to randomly distribute block mirrors throughout the entire array. This approach allows XIV to reconstruct the data on very large 2 TB HDDs in less than 30 minutes.
RAID + Transformation:
Self-healing storage and BeyondRAID are two examples of transformation.
In Self- healing, HDD faults are reduced by proactively healing hard disk drives before a fault occurs using HDD reconditioning algorithms. It also uses advanced vibration controls and sealed systems called DataPacs to reduce outside influences from causing HDD faults. Some examples: Atrato Inc.’s Velocity1000 (V1000), DataDirect Networks Inc.’s DDN S2A technology, Panasas Inc.’s ActiveScan technology, 3PAR’s InSpire Architecture.
BeyondRAID chooses RAID sets based on the required data protection at any given point in time. It protects against one or two HDD failures and has built-in automatic data self-healing (not storage self-healing). Data blocks are spread across all of the drives so data reconstruction is very fast. Because the system is “data aware,” it allows for different drive sizes, drive re-ordering and proportional rebuild times.
BeyondRAID Explained (YouTube video)
RAID traditional levels: RAID, 0, 1, 2, 10, 01, 3, 3 double parity, 4, 5, 6, 50, 60
Links:
May2010 Storage Magazine (You may need account to access it):
http://viewer.media.bitpipe.com/1127859972_677/1273085309_971/StoragemagOnlineMay2010.pdf
http://viewer.media.bitpipe.com/1127859972_677/1273085309_971/StoragemagOnlineMay2010.pdf
How to setup RAID on your PC:
http://www.pcworld.com/businesscenter/article/132877/how_to_set_up_raid_on_your_pc.html
Hitachi is launching 4 TB HDD:
http://www.itproportal.com/2011/09/08/hitachi-launches-4tb-hard-drive/
http://www.itproportal.com/2011/09/08/hitachi-launches-4tb-hard-drive/
Great posting.
ReplyDelete