A Case for Redundant Arrays of Inexpensive Disks…

The RAID paper asks how many small disks can deliver more I/O without making storage collapse whenever one device fails. Read it to see how striping exposes parallelism, how mirrors or XOR parity make one missing disk reconstructable, and why rotating the parity position removes a dedicated check-disk bottleneck without proving modern latency or whole-system durability.

Reading focus: Why an unprotected array encounters a disk failure sooner even as more independent devices increase potential I/O bandwidth. How one parity bit per stripe lets the array reconstruct any single known missing bit from the survivors. Why RAID levels trade storage overhead, small-write work, and parallelism, and which failure assumptions bound the paper's reliability arithmetic.

ACM SIGMOD 1988. David A. Patterson, Garth A. Gibson, and Randy H. Katz. 30 min read, easy difficulty.