When we talk about data storage and protection, the term 'RAID' often pops up. It's a clever way to combine multiple hard drives into a single logical unit, offering benefits like improved performance or redundancy against drive failures. Today, I want to dive into two specific RAID levels that often get mentioned together: RAID 50 and RAID 60. They sound similar, and in a way, they are, but understanding their nuances can be really helpful.
Think of RAID 50 and RAID 60 as 'nested' RAID levels. They don't stand alone like RAID 0 or RAID 1. Instead, they build upon other RAID configurations. Specifically, they combine the strengths of RAID 0 (striping for performance) with either RAID 5 or RAID 6 (parity for redundancy).
Let's break it down. RAID 50, for instance, is essentially a RAID 0 array made up of multiple RAID 5 'spans'. A span is just a group of drives working together. So, you have several RAID 5 groups, and then those groups are striped together using RAID 0. This gives you the best of both worlds: the fault tolerance of RAID 5 within each span, and the speed boost from striping across those spans.
Now, RAID 60 takes this a step further. It's a RAID 0 array composed of multiple RAID 6 'spans'. RAID 6 is like RAID 5 but with double the parity information. This means it can withstand the failure of two drives within a single span without losing data. When you combine this robust redundancy with the striping of RAID 0, you get RAID 60. This level offers a very high degree of data protection, making it suitable for critical applications where downtime and data loss are simply not an option.
Looking at the technical specs, the Avago SAS3508 controller, for example, can support a good number of drives, and these nested RAID levels leverage that. For RAID 50 and 60, the number of drives per group and the number of virtual drives are key. You're looking at drive groups that can hold a significant number of drives, and within those, you can define multiple virtual drives. The reference material mentions that each span in a RAID 10 or 50 array allows for one failed drive, which makes sense because the underlying RAID 5 span can handle one failure. For RAID 60, each span can tolerate up to two failed drives, reflecting the underlying RAID 6 protection.
It's interesting to see how these levels are designed. The total number of drives supported is quite large, and this is the sum of active drives, hot spares, and those in an 'Unconfigured Good' state, ready to be brought into an array. The flexibility to create many virtual drives and drive groups means you can tailor your storage to specific needs.
When you compare them, the main difference boils down to the level of redundancy. RAID 50 is great for performance and can handle a single drive failure within its spans. RAID 60, by using RAID 6 spans, offers a higher level of protection, capable of surviving two drive failures per span. This extra layer of safety in RAID 60 comes at a slight cost, potentially in performance or complexity, but for many high-availability scenarios, it's a worthwhile trade-off.
It's a bit like choosing between a reliable car and a super-secure armored vehicle. Both get you there, but one offers significantly more protection against unexpected events. Understanding these differences helps in making informed decisions about how to best safeguard your valuable data.
