Enterprise Backup and Storage

Prevention of data loss is an ongoing concern for all IT administrators. Whether an accidental deletion of a file, a corruption of a database, or the functional loss of a server or the data center, plans and processes need to be in place to quickly recover a customer’s environment to a full production status. In order to better support our customer’s needs for backup and disaster recovery (DR) processes, Texas A&M IT has implemented IBM’s Tivoli Storage Manager (TSM), a LAN-based data management product. Texas A&M IT's Business Continuance (BC) Team is assigned to manage and monitor these processes.

Methodology

TSM uses a “backup everything” methodology when processing a specific server node. All data accessible (disk, CD, NFS, etc.) is included in a backup scan unless specifically excluded. TSM uses a robust INCLUDE/EXCLUDE rule set for defining a customer's backup needs. Charging for the TSM backup/DR service is based on total data storage costs (total GB stored).

As originally designed and scaled, TSM's main function is for DR processes, i.e., the timely recovery of a customer’s environment from a point-in-time to full production status. Long-term retention of data (backups) is not generally encouraged, but can be supported. Long-term data migration (commonly called archiving) is not supported past one year due to resource contention. Archiving services may be offered in the future if deemed necessary for customers.

Backups

Texas A&M IT uses TSM in a “continuous incremental” mode for creating backup and DR copies of customer’s data. To protect the data from incorrect classification, all data processed through TSM is treated as “sensitive” and is encrypted in-flight between client server and TSM server. For each backup cycle (usually 24 hours), changed files are identified (by comparing to last version) and copied over the LAN to the TSM backup server.  Concurrently, data is replicated to a co-located data center in Dallas (Databank).

The BC Team works with each customer to set up their backup environment to their needs. Though most settings are customizable, the following settings are default:

  • The BC Team initiates and manages the backup/restore processes on the customer's server(s)
  • The customer initiates and manages the backup/restore of databases.  BC Team provides assistance to DBAs as needed.
  • File level backups done nightly between 6:30 p.m. and 6:00 a.m.
  • Retention for copies of each data file set to 14 versions (not days).
  • Retention of expired data (rolled-off version) is 30 days
  • The BC Team works one-on-one with the customer to customize inclusion/exclusion lists for volatile and unnecessary data (log files, networked drives, peripherals, etc.)

Recovery

Depending on the type of recovery required (single file, entire system), the BC Team can assist customers in getting their system back to full working order. Generally, restore requests are made, prioritized and tracked through the ServiceNow system, though service “Enterprise Backup - TSM (TAMU)”. Also, an option for customer-initiated restores is available.