Next-Generation, Non-Disruptive, Dataset-Level Migration  

As information flows across storage arrays in ever-growing chunks, the demand for better performance and response times grows exponentially. IT storage administrators are looking to take advantage of today’s new and improved storage arrays and larger disk drives and volumes. They also need to reduce TCO by managing smaller number of volumes and to minimize system constraints such as Unit Control Block (UCB) and floor space limitations. 

To reduce these constraints, IT storage administrators are consolidating from smaller to larger capacity volumes (i.e., from MOD3 to MOD9 to MOD27). Furthermore, storage administrators move datasets to fewer disks or more expensive disks based on application activities and service level agreements in order to better optimize their performance and implement tier storage within the array. 

The migration process presents a major hurdle: the need to plan and prepare for application downtime—and to ensure the proper resources are onsite for the duration of the migration. Due to downtime limitations, migrations often must be performed over several weekends. Additionally, the manual migration process is error-prone and problems often arise in the form of unexpected results, such as insufficient disk space on target volumes, data corruption, the inability to bring an application online after the migration is completed, and other errors caused by the manual process. 

In recent years, new tools have become available that allow administrators to non-disruptively move volumes and migrate data without taking their applications offline. However, those tools only permit migration of full volumes and do not allow the non-disruptive migration of individual datasets for performance tuning and volume consolidation. 

Non-Disruptive Dataset-Level Migration  

Enter Logical Data Migration Facility (LDMF) from EMC/Softek. Unique to the marketplace, LDMF is the only solution that allows the non-disruptive migration of datasets between unlike size volumes and any storage array. In addition, LDMF facilitates that process with absolutely no application downtime. 

LDMF Key Functionality and Benefits  

Non-disruptive migration with zero downtime: Once activated, LDMF copies the selected datasets from the source to the target volumes while the application remains available to users. Storage administrators can review via the ISPF screens the percentage of data copied to the new location. Once all the data has been copied over, LDMF mirrors each I/O and writes it to both the source and target locations simultaneously. This process guarantees both source and target copies are identical—until the storage administrator chooses to switch over to the new location. 

At this point, LDMF modifies the ICF catalog entries to point to the new location of the datasets. From that moment on, all I/O is redirected to the new location on the target side only and the migration process is complete. The entire migration process is completed without taking the application offline even for one second. 

Selectable datasets using ISPF screens: LDMF facilitates the use of ISPF screens to easily select the source datasets using wild cards. Target volumes can be selected simply by specifying SMS group or specific volumes. 

Safer Dataset-Level Migration: LDMF helps prevent many errors such as insufficient disk space on target volumes. Prior to the migration process, LDMF automatically ensures that target volumes have sufficient disk space to perform the migration. Furthermore, customers can be sure all the datasets selected are migrated to the target volumes. 

Powerful solution for dynamic needs: In the final analysis, LDMF provides a powerful and cost-effective dataset-level migration solution that directly addresses the dynamic needs of its enterprise customers. With LDMF, storage administrators need not perform data migrations over weekends—and there’s no need to schedule application downtime. The entire migration process can be performed while the application is up and running.