Fogo Server Certification Practice Test

Redundancy and how write work differ between these two RAID levels. RAID 1 keeps an exact copy of all data on two disks, so you get complete redundancy—if one disk fails, the other has your data. Because there are two copies, read requests can be served from either disk, which often speeds up reads. But every write must be performed on both disks, so the write overhead is higher and write throughput is reduced compared with a single disk.

RAID 5 stripes data across multiple disks and adds parity information. This gives redundancy with much better storage efficiency than a full mirror, since you don’t replicate all data—usable capacity is reduced by only one disk’s worth of parity. Reads can be fast because data is spread across disks, so multiple reads can happen in parallel. Writes, however, incur parity calculations: when you update data, the system must read the old data and old parity, compute new parity, and then write both the new data and the new parity. That extra parity work makes writes slower and more I/O-intensive than RAID 1 in many scenarios.

So the best description is that RAID 1 provides redundancy via mirroring with improved reads but higher write overhead, while RAID 5 provides redundancy with better storage efficiency but heavier write parity calculations. The idea that RAID 1 is faster for writes or that RAID 5 is worse for reads isn’t generally correct, since write overhead and read performance depend on workload and array size, but the parity-based write penalty in RAID 5 is a key differentiator.