These exemplary erasure code architectures may further be improved by providing protection against failures in a system in which each storage unit has regions with systematically higher failure rates than other regions on the same storage unit. In other words, by identifying regions on a given storage unit which have a higher probability of experiencing an error, each erasure code strip on a given storage unit (e.g., optical disk) may be mapped such that its containing erasure code stripe includes only one strip containing high failure rate regions, in addition to strips from lower failure rate regions on other storage units. As a result, error recovery code failures are significantly reduced, e.g., as will be described in further detail below with respect to method 500. For instance, operation 502 of method 500 includes selecting strips from each optical disk for the given erasure code stripe such that the given erasure code stripe includes at most one strip from a high failure rate region of the respective optical disk, e.g., as will be described in further detail below.