6.4. Basic Reliability Modelling

With the help of data replication or parity coding, the reliability of a disk array can be greatly improved. Several studies have quantified the reliability of homogeneous disk arrays with mirroring [115], parity coding schemes (RAID) [128], [117], [78], [30], [13], or a combination of both [14].

The primary metric to compare the reliability of multi-disk storage systems is typically the mean time to failure (MTTF) of a system. The most critical failures typically involve loss of data. Hence in such situations, the MTTF is rephrased as the mean time to data loss (MTTDL). This would be a metric for traditional transaction oriented applications, for example. For a SM server, however, we assume an architecture with ...

Get Streaming Media Server Design now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.