Abstract: A distributed network of storage elements (DNSE) is provided in which the physical capacity of each drive is split into a set of equal sized logical splits which are individually protected within the DNSE using separate RAID groups. To reduce restoration latency, members of the RAID groups having a member in common on a given drive are spread within the DNSE to minimize the number of sets of drives within the DNSE that have RAID members in common. By causing the splits to be protected by RAID groups, restoration of the splits may occur in parallel involving multiple drives within the DNSE. By minimizing the overlap between RAID members on various drives, failure of a given drive will not require multiple reads from another drive in the DNSE. Likewise, spare splits are distributed to enable write recovery to be performed in parallel on multiple drives within the DNSE.

Abstract: Configuring spare sections in a system having a plurality of devices and having a plurality of erasure encoding groups, each of the devices having a plurality of sections and each of the groups having a plurality of members located on the sections of the devices, includes configuring the system to have a plurality of erasure encoding groups with no spare sections, ordering the groups according to desirability of converting members of each of the groups into spare sections, converting members of a subset of the groups into spare sections according to ordering of the groups, and determining if the spare sections for the system provide sufficient coverage for the system in response to failure of each of the devices. The erasure encoding may be a RAID encoding.

Abstract: Individual storage devices of a RAID group are monitored for faults. A health indicator for each storage device is calculated based on fault growth rate. Non-failed storage device are swapped out based on the health indicator. Techniques for monitoring the storage devices include background media scans and growth list polling.

Abstract: A distributed network of storage elements (DNSE) is provided in which the physical capacity of each drive is split into a set of equal sized logical splits which are individually protected within the DNSE using separate RAID groups. To reduce restoration latency, members of the RAID groups having a member in common on a given drive are spread within the DNSE to minimize the number of sets of drives within the DNSE that have RAID members in common. By causing the splits to be protected by RAID groups, restoration of the splits may occur in parallel involving multiple drives within the DNSE. By minimizing the overlap between RAID members on various drives, failure of a given drive will not require multiple reads from another drive in the DNSE. Likewise, spare splits are distributed to enable write recovery to be performed in parallel on multiple drives within the DNSE.

Abstract: Embodiments are directed to techniques for techniques for selecting a proper set of spare sections to use in a given failure scenario. These embodiments use a set of rules to define which spare sections are eligible to serve as spares for reconstruction of the RAID members on a disk that had failed. In addition, the set of rules may include weighted rules to allow optimization in the spare selection process.

Abstract: Configuring spare sections in a system having a plurality of devices and having a plurality of erasure encoding groups, each of the devices having a plurality of sections and each of the groups having a plurality of members located on the sections of the devices, includes configuring the system to have a plurality of erasure encoding groups with no spare sections, ordering the groups according to desirability of converting members of each of the groups into spare sections, converting members of a subset of the groups into spare sections according to ordering of the groups, and determining if the spare sections for the system provide sufficient coverage for the system in response to failure of each of the devices. The erasure encoding may be a RAID encoding.

Abstract: Described are techniques for selecting a storage pool configuration. A set of one or more candidate storage pool configurations is determined. For each of the candidate storage pool configurations of the set, a reliability calculation is performed to determine a reliability value indicating reliability of the candidate storage pool configuration. For each of the candidate storage pool configurations of the set, an availability calculation is performed to determine an availability value indicating availability of the candidate storage pool configuration. In accordance with selection criteria, one of the candidate storage pool configurations of the set for a first storage pool is selected. The selection criteria includes the availability value and the reliability value for each of the plurality of candidate storage pool configurations.

Abstract: A decision support system, including a decision model, provides an indication, either of itself or in combination with a weighted average of the scores assigned to the decision model factors, of the degree to which the pattern of contribution scores corresponds to a benchmark pattern. This indication, referred to as a matching index, can be calculated according to a class of matching index functions, including those based on regression-correlation and Bayesian pattern matching and cost/loss concepts. In addition, the distribution quality and bias can be indicated by another measure referred to as the distribution quality bias measure which is an indication of how biased such a contribution distribution is, with respect to a benchmark of contributions. The matching index and distribution quality and bias measure can be combined to indicate an overall quality value of match, as opposed to a simple measure of pattern matching or weighted averages.

Abstract: A decision support system, including a decision model, provides an indication, either of itself or in combination with a weighted average of the scores assigned to the decision model factors, of the degree to which the pattern of contribution scores corresponds to a benchmark pattern. This indication, referred to as a matching index, can be calculated according to a class of matching index functions, including those based on regression-correlation and Bayesian pattern matching and cost/loss concepts. The matching index also allows the designation of model factors as `Must Have` or `Must Not Have` factors, and asymmetric functions for special circumstances. The decision support system can be used to assist in the control of a variety of systems and/or processes and/or in the analysis of information, wherever patterns of information may assist in the selection of an alternative from amongst a set of decision alternatives.

Abstract: A method and system for balancing weights in models for decision support systems. When weights in such a model are assigned as local weights, evaluators may obtain non-intuitive results due to biases introduced by variations in the number of child factors of parent factors at a given level. Adjusting weights assigned within the model provides results which are often more intuitive to an evaluator. Two methods of calculating appropriate balancing coefficients to adjust weights are presented.