05-28-201004:01 AM
Hopefully this helps -
Network RAID 5 - Stores three data and one parity as a minimum configuration i.e 3+1 , meaning four nodes as a starting point.
Network RAID 6 - Stores four data and two parity as a minimum configuration i.e 4+2 , meaning six nodes are required initially.
Network RAID 10 - Stores two copies of each volumea s block, providing continuous data availability across any single node failure.
Network RAID 10+1 - Stores three copies of each block for mission-critical data that needs to be available despite any double node failure.
Network RAID 10+2 - used in situations where a cluster is divided between two locations, and the data must be continuously available in the event of both a site failure and a node failure at the alternate site (Campus SAN).
While the resiliency of the network raid flavors may be similar from one to another. How they are managed, used, and the workloads they are suited to is the real difference. Supposedly Network RAID 5 and 6 provide "space utilization benefits", but the use of snapshots is required to see the benefits (according to the HP/LHN docs). Even HP/LHN suggests RAID5/6 should be used on mostly read, sequential workloads on larger clusters, such as file shares and archiving. Whereas RAID 10+X is more appropriate for highly transactional workloads.
I'm personally sticking with NW RAID-10+x as I've yet to see any real examples of significant storage savings using NW RAID5/6.
Network RAID 5 - Stores three data and one parity as a minimum configuration i.e 3+1 , meaning four nodes as a starting point.
Network RAID 6 - Stores four data and two parity as a minimum configuration i.e 4+2 , meaning six nodes are required initially.
Network RAID 10 - Stores two copies of each volumea s block, providing continuous data availability across any single node failure.
Network RAID 10+1 - Stores three copies of each block for mission-critical data that needs to be available despite any double node failure.
Network RAID 10+2 - used in situations where a cluster is divided between two locations, and the data must be continuously available in the event of both a site failure and a node failure at the alternate site (Campus SAN).
While the resiliency of the network raid flavors may be similar from one to another. How they are managed, used, and the workloads they are suited to is the real difference. Supposedly Network RAID 5 and 6 provide "space utilization benefits", but the use of snapshots is required to see the benefits (according to the HP/LHN docs). Even HP/LHN suggests RAID5/6 should be used on mostly read, sequential workloads on larger clusters, such as file shares and archiving. Whereas RAID 10+X is more appropriate for highly transactional workloads.
I'm personally sticking with NW RAID-10+x as I've yet to see any real examples of significant storage savings using NW RAID5/6.
Hope that clears things up more.
参考链接：
本文转自daniel8294 51CTO博客，原文链接：http://blog.51cto.com/acadia627/1337534，如需转载请自行联系原作者
