Cloud Connector Processing
Cloud Connector (CUZ) backs up any sequential data from tape or disk to the cloud and also restores it to DASD or tape. You define the cloud servers, staging options, backup filter criteria, data restore settings, and other general options in a parameter member. Cloud Connector uses the options in the parameter member to connect to the cloud servers that you specified. You can set up staging options so that the data is copied ("staged") to DASD before going to the cloud. This ensures that a copy of the data exists in case there is a connection issue with the cloud.
The data sets selected by your backup filter criteria will be saved to the cloud. You can set a retention period to determine how long the data sets will be stored in the cloud. You can display a list of the data sets that have been written to the cloud and can filter by data set name or cloud server.
When you restore a data set from the cloud, it will be written to either DASD or tape, based on the restore options you specify. Data sets with block sizes over 32K must be restored to tape.
There are two methods for backing up data to the cloud:
- Capture data while it is being written to tape or DASD and copy to the cloud. This is done through filter criteria defined in the parameter member, CUZ#PARM. You can filter by data set name, esoteric unit, or SMS storage class.
- Create a list of existing data sets (history data sets) to copy directly to the cloud. This method allows you to locate already existing sequential files and copy them directly to the cloud. The pre-existing data sets are identified by name in parameter member, CUZ#INEX.
- Staging data to DASD. This option will quickly save the data to a DASD data set called a
staging file. After the batch job has completed, the contents of the staging file are written
to the cloud by the Cloud Connector Started Task. This method is
recommended for two reasons:
- Writing data directly to the cloud requires an uninterrupted transmission via a stable network connection to the cloud. If there is any interruption in that transmission, the data will not be written to the cloud.
- If the data is staged to DASD, and the network connection is interrupted, the process to copy that data to the cloud will continue to restart until the data has been completely written to the cloud.
- Writing data directly to the cloud. This option will capture a copy of the data you are writing and send it to the cloud concurrently while the batch job executes. This option works best when transmission is uninterrupted and a stable network connection to the cloud is available. Writing data directly to the cloud could, in some cases, increase the elapsed times of your batch jobs if there is network delay.
We recommend using the staging method due to the ability to restart the copy process. If the staging method is used, the data will be written to the cloud even if the copy process must be restarted to accomplish this task. If the "Direct Cloud Write" method is used, any interruptions in transmission will result in failure to write the data to the cloud.
Cloud Connector has a started task with the default name of CUZCLOUD, which coordinates the movement of files to the cloud. Parameter member CUZ#PARM enables you to define the files you want to copy to the cloud based on data set name, esoteric unit, or SMS storage class. Another parameter member, CUZ#INEX, enables you to copy pre-existing files ("history files") to the cloud based on data set name.
The CUZCLOUD started task accesses one or more parameter data sets that contain the CUZ#PARM member (for the filter criteria and cloud definitions), and the CUZ#INEX member (for the data set names or masks for pre-existing files). The CUZCLOUD started task must be active in order for data sets to be written to or restored from the cloud.