1. A method of designing a storage system comprising:
- assigning decision variables to a range of candidate storage systems, each of the decision variables identifying a single candidate storage system; and
- employing, by a computer, the decision variables in a mathematical program to determine which of the candidate storage systems meets availability and reliability criteria and incurs a near minimal projected cost, wherein the availability criterion is a measure of an ability to access data when desired, and the reliability criterion is one of a measure of absence of data loss and an indicator of an acceptable amount of data loss.
2. The method of claim 1 wherein the near minimal projected cost comprises a minimal projected cost.
3. The method of claim 1 wherein each of the candidate storage systems comprises a primary storage and a secondary storage, wherein the mathematical program has input values specifying different types of the secondary storage.
4. The method of claim 3 wherein the secondary storage is selected from a range of remote mirror configurations.
5. The method of claim 4 wherein each of the range of remote mirror configurations comprises secondary disk storage and a network bandwidth between the primary storage and the secondary disk storage.
6. The method of claim 5 wherein at least some of the range of remote mirror configurations further comprise window durations for batched asynchronous mirroring.
7. The method of claim 3 wherein the secondary storage is selected from a range of backup configurations.
8. The method of claim 7 wherein at least some of the range of backup configurations comprise local backups with a remote vault component.
9. The method of claim 7 wherein at least some of the backup configurations each comprise full backups.
10. The method of claim 7 wherein at least some of the backup configurations include incremental backups.
11. The method of claim 7 wherein at least some of the backup configurations comprise a quantity of tape drives and a quantity of tapes.
12. The method of claim 11 wherein at least some of the backup configurations comprise one or more tape libraries.
13. The method of claim 3 wherein the secondary storage is selected from a range of remote mirror configurations and backup configurations.
14. The method of claim 1 wherein the availability and reliability criteria comprise a recovery time objective and a recovery point objective, respectively.
15. The method of claim 1 wherein the mathematical program comprises a mixed integer program which comprises constraints and an objective, and wherein the mixed integer program comprises integer variables and continuous variables.
16. The method of claim 15 wherein the constraints comprise:
- selecting one and only one of the candidate storage systems, thereby selecting a particular candidate storage system;
- selecting one and only one spare resource option;
- satisfying a bandwidth constraint;
- satisfying a reload constraint;
- satisfying a recovery time constraint; and
- satisfying a recovery point constraint.
17. The method of claim 16 wherein the objective comprises minimizing an outlay cost.
18. The method of claim 16 wherein the objective comprises minimizing a sum of an outlay cost and a penalty cost.
19. The method of claim 17 wherein the outlay cost comprises a sum of costs for a secondary storage and the spare resource option.
20. The method of claim 17 further comprising the step of iteratively instantiating the mathematical program to map at least a portion of a solution space.
21. The method of claim 19 wherein the sum of costs further comprises a cost for a primary storage.
22. The method of claim 19 wherein the secondary storage comprises a remote mirror.
23. The method of claim 19 wherein the secondary storage comprise: a backup configuration.
24. The method of claim 23 wherein the backup configuration comprises a tape backup.
25. The method of claim 24 wherein the bandwidth constraint comprises ensuring that a quantity of tape drives and a quantity of available tapes satisfy a storage workload.
26. The method of claim 21 wherein the primary storage comprises one or more disk arrays.
27. A method of designing a storage system comprising:
- assigning decision variables to a range of candidate storage systems, each of the decision variables identifying a single candidate storage system; and
- employing, by a computer, the decision variables in a mathematical program to determine which of the candidate storage systems meets availability and reliability criteria and incurs a near minimal projected cost,
- wherein the mathematical program comprises a mixed integer program which comprises constraints and an objective,
- wherein the constraints comprise:
- selecting one and only one of the candidate storage systems, thereby selecting a particular candidate storage system,
- selecting one and only one spare resource option,
- satisfying a bandwidth constraint,
- satisfying a reload constraint,
- satisfying a recovery time constraint, and
- satisfying a recovery point constraint,
- wherein the objective comprises minimizing an outlay cost, wherein the outlay cost comprises a sum of costs for a secondary storage and the spare resource option, wherein the secondary storage comprises a remote mirror, and
- wherein the bandwidth constraint comprises ensuring that a quantity of network links between a primary storage and the secondary storage satisfies a storage workload.
28. The method of claim 27 wherein the reload constraint ensures that a bandwidth for the quantity of network links between the primary storage and the secondary storage does not exceed a reload rate for the primary storage.
29. A method of designing a storage system comprising:
- assigning decision variables to a range of candidate storage systems, each of the decision variables identifying a single candidate storage system; and
- employing, by a computer, the decision variables in a mathematical program to determine which of the candidate storage systems meets availability and reliability criteria and incurs a near minimal projected cost,
- wherein the mathematical program comprises a mixed integer program which comprises constraints and an objective,
- wherein the constraints comprise:
- selecting one and only one of the candidate storage systems, thereby selecting a particular candidate storage system,
- selecting one and only one spare resource option,
- satisfying a bandwidth constraint,
- satisfying a reload constraint,
- satisfying a recovery time constraint, and
- satisfying a recovery point constraint,
- wherein the objective comprises minimizing an outlay cost, wherein the outlay cost comprises a sum of costs for a secondary storage and the spare resource option. wherein the secondary storage comprises a backup configuration, wherein the backup configuration comprises a tape backup,
- wherein the bandwidth constraint comprises ensuring that a quantity of tape drives and a quantity of available tapes satisfy a storage workload, and
- wherein the reload constraint ensures that a bandwidth for the quantity of tape drives does not exceed a reload rate for a primary storage.
30. A method of designing a storage system comprising:
- assigning decision variables to a range of candidate storage systems, each of the decision variables identifying a single candidate storage system; and
- employing, by a computer, the decision variables in a mathematical program to determine which of the candidate storage systems meets availability and reliability criteria and incurs a near minimal projected cost,
- wherein the mathematical program comprises a mixed integer program which comprises constraints and an objective,
- wherein the constraints comprise:
- selecting one and only one of the candidate storage systems, thereby selecting a particular candidate storage system,
- selecting one and only one spare resource option,
- satisfying a bandwidth constraint,
- satisfying a reload constraint,
- satisfying a recovery time constraint, and
- satisfying a recovery point constraint,
- wherein the objective comprises minimizing a sum of an outlay cost and a penalty cost, and
- wherein the recovery time constraint comprises ensuring that a recovery time minus a recovery time allowance does not exceed a recovery time objective.
31. The method of claim 30 wherein the recovery point constraint comprises ensuring that a recovery point minus a recovery point allowance does not exceed a recovery point objective.
32. The method of claim 31 wherein the penalty cost comprises a penalty rate function.
33. The method of claim 31 wherein the penalty cost comprises a likelihood of failure times a sum of a recovery time penalty rate times the recovery time allowance and a recovery point penalty rate times the recovery point allowance.
34. The method of claim 31 wherein the penalty cost comprises a sum of a recovery time penalty rate times the recovery time allowance and a recovery point penalty rate times the recovery point allowance.
35. The method of claim 31 wherein the recovery time objective is set to zero or the recovery point objective is set to zero or both the recovery time objective and the recovery point objective are set to zero in an instantiation of the method of designing the storage system.
36. A method of designing a storage system comprising:
- assigning a decision variable to each of a range of candidate storage systems; and
- employing, by a computer, the decision variables in a mixed integer program which
- comprises an objective and constraints, the objective comprises minimizing a projected cost and the constraints comprise:
- selecting one the candidate storage systems, thereby selecting a particular candidate storage system;
- satisfying a bandwidth constraint;
- satisfying a recovery time constraint, wherein the recovery time constraint specifies a time period for recovery from a failure; and
- satisfying a recovery point constraint, wherein the recovery point constraint specifies a worst case time period for data loss.
37. The method of claim 36 wherein the projected cost comprises an outlay cost.
38. The method of claim 37 wherein the projected cost further comprises a penalty cost.
39. The method of claim 36, wherein the mixed integer program has inputs related to failure scenarios that include a scope of failure and a failure likelihood.
40. A computer readable media comprising computer code that upon execution by a computer implements a method of designing a storage system, the method of designing the storage system comprising:
- assigning a decision variable to each of a range of candidate storage systems; and
- employing the decision variables in a mathematical program to determine which of the candidate storage systems meets availability and reliability criteria and incurs a near minimal projected cost, wherein the availability criterion is a measure of an ability to access data when desired, and the reliability criterion is one of a measure of absence of data loss and an indicator of an acceptable amount of data loss.
41. A computer readable media comprising computer code that upon execution by a computer implements a method of designing a storage system, the method of designing the storage system comprising:
- assigning a decision variable to each of a range of candidate storage systems; and
- employing the decision variables in a mixed integer program which comprises an objective and constraints, the objective comprises minimizing a projected cost and the constraints comprise:
- selecting one of the candidate storage systems, thereby selecting a particular candidate storage system;
- satisfying a bandwidth constraint;
- satisfying a recovery time constraint, wherein the recovery time constraint specifies a time period for recovery from a failure; and
- satisfying a recovery point constraint, wherein the recovery point constraint specifies a worst case time period for data loss.
