Optimization of disk system read-ahead processes
Most advanced disk systems can automatically optimize the performance of read operations when the disk system can detect sequential reads. When the disk system detects sequential reads, it can have the data for the next read in cache, or at least have the read in progress.
Disk systems detect sequential reads on a LUN-by-LUN basis. However, a sequential read might not be detected if more than one read is in progress for the same LUN. Disk systems do not perceive file systems or files within a LUN and distinguish only the blocks that are being accessed. When two sequential reads are in progress on one LUN, the blocks that are being accessed no longer respond as sequential blocks. The blocks appear to come from different places, and the read-ahead optimizations are typically stopped.