Abstract: Restoration devices in a cloud storage system are paired with source containers associated with a mainframe computer, and series of commands are generated based on the pairings to cause copies of data at the source containers to be stored to the restoration devices. A point-in-time copy of the copy of the data at the source containers may be stored to some restoration devices, and a second copy of the data may be stored to other restoration devices. The restoration devices may be reallocated from inactive source containers. Execution of the commands is monitored, and the commands are modified if the execution of the commands does not satisfy one or more desired conditions. For example, a cycle time associated with copying data to a restoration device may be measured, and if the cycle time exceeds a threshold, the command may be modified.

Abstract: Restoration devices in a cloud storage system are paired with source containers associated with a mainframe computer, and series of commands are generated based on the pairings to cause copies of data at the source containers to be stored to the restoration devices. A point-in-time copy of the copy of the data at the source containers may be stored to some restoration devices, and a second copy of the data may be stored to other restoration devices. The restoration devices may be reallocated from inactive source containers. Execution of the commands is monitored, and the commands are modified if the execution of the commands does not satisfy one or more desired conditions. For example, a cycle time associated with copying data to a restoration device may be measured, and if the cycle time exceeds a threshold, the command may be modified.

Abstract: A method of forecasting having the steps of extracting a number of computer usage records and storing the computer usage records in a records file (102). The computer usage records are then sorted according to date (104). The computer usage records are summarized within a number of date ranges to form a usage summarization record (106). A curve fitting model is applied to the usage summarization record to predict a computer usage trend curve (108). A computer usage forecast is then input (110). A computer usage growth curve is based on the computer usage forecast (112). The computer usage trend curve and the computer usage growth curve are displayed (114).