You can specify that your member operate with a single checkpoint data set. In most cases, this mode is not recommended. Effectively, you are processing in an environment in which there will be no backup copy of the checkpoint data set. As a result, JES2 will not need to read/write to two data sets. If that data set becomes damaged, lost, or suffers an I/O error, and your member experiences a concurrent failure, you will have no checkpoint data set available to restart JES2.

However, depending on your specific volume or structure use and configuration, it might not be practical for particular members of your JES2 configuration to maintain a backup checkpoint. For example, a 32-way MAS could be composed of two 3090 processors that manage a heavy batch workload and 30 smaller processors that manage on-line transaction programs. To reduce the cost of storage overhead, this installation might run without a backup checkpoint on the 30 smaller processors, while ensuring that the 3090s operate with DUPLEX=YES.

Another scenario could involve a MAS composed of different level processors. An installation can determine that it is not efficient to run with a backup checkpoint on the slower processors.