Maintenance cost estimation

1. A method of providing maintenance to a storage system, comprising:

initializing weights of neurons of a plurality of self-organized map neural network models that output time and materials estimates based on input configuration data, wherein a separate model is constructed for each type of maintenance operation that may be performed on the storage system;

providing training data that is used to adjust the weights of each of neurons of each of the self-organized map neural network models;

providing configuration data of the storage system to the one of the self-organized map neural network models corresponding to the type of maintenance operation being performed;

obtaining an estimate of maintenance time and materials needed for the maintenance operation based on the configuration data provided to the one of the models corresponding to the type of maintenance operation being performed;

performing the maintenance; and

providing actual maintenance parameters resulting from performing the maintenance as additional training data for the one of the models corresponding to the type of maintenance operation being performed.

2. A method, according to claim 1 , wherein weights of neurons of the self-organized map are initialized randomly.

3. A method, according to claim 1 , wherein the training data includes at least one of: an I/O load of the storage system, memory size of the storage system, a drive count of the storage system, and size and parameter information corresponding to hardware being added for the maintenance operation.

4. A method, according to claim 3 , wherein the size and parameter information corresponding to hardware being added includes at least one of: physical storage unit capacity of the hardware, a CPU count of the hardware, and a memory size of the hardware.

5. A method, according to claim 1 , wherein the training data includes actual time and materials for prior storage system maintenance operations used for the training data.

6. A method, according to claim 1 , wherein the estimate of maintenance time and materials is broken into separate phases.

7. A method, according to claim 1 , wherein the model is provided on the storage system.

8. A method, according to claim 1 , wherein at least one of the models corresponds to an online engine add maintenance procedure in which the storage system is running while more storage capacity is added to the storage system.

9. A non-transitory computer readable medium containing software that facilitates providing maintenance to a storage system, the software comprising:

executable code that initializes weights of neurons of a plurality of self-organized map neural network models that output time and materials estimates based on input configuration data, wherein a separate model is constructed for each type of maintenance operation that may be performed on the storage system and wherein training data that is used to adjust the weights of each of neurons of each of the self-organized map neural network models;

executable code that provides configuration data of the storage system to the one of the self-organized map neural network models corresponding to the type of maintenance operation being performed;

executable code that obtains an estimate of maintenance time and materials needed for the maintenance operation based on the configuration data provided to the one of the self-organized map neural network models corresponding to the type of maintenance operation being performed; and

executable code that provides actual maintenance parameters resulting from performing the maintenance as additional training data for the one of the models corresponding to the type of maintenance operation being performed.

10. A non-transitory computer readable medium, according to claim 9 , wherein weights of neurons of the self-organized map are initialized randomly.

11. A non-transitory computer readable medium, according to claim 9 , wherein the training data includes at least one of: an I/O load of the storage system, memory size of the storage system, a drive count of the storage system, and size and parameter information corresponding to hardware being added for the maintenance operation.

12. A non-transitory computer readable medium, according to claim 11 , wherein the size and parameter information corresponding to hardware being added includes at least one of: physical storage unit capacity of the hardware, a CPU count of the hardware, and a memory size of the hardware.

13. A non-transitory computer readable medium, according to claim 9 , wherein the training data includes actual time and materials for prior storage system maintenance operations used for the training data.

14. A non-transitory computer readable medium, according to claim 9 , wherein the estimate of maintenance time and materials is broken into separate phases.

15. A non-transitory computer readable medium, according to claim 9 , wherein the software is provided on the storage system.

16. A non-transitory computer readable medium, according to claim 9 , wherein at least one of the models corresponds to an online engine add maintenance procedure in which the storage system is running while more storage capacity is added to the storage system.