Application Management and Support

System Management

The System Management team is responsible for keeping the HP NonStop servers running efficiently . They are responsible for all aspects of hardware, software and communications that support the enterprise applications.

The System Manager should notify the Security Administration when changes, upgrades or other events could affect the security of the enterprise application.

The System Management methods must also comply with the Corporate Security Policy and Standards, so a good relationship between the systems team and security team is vital .

Database Management

The database management team is responsible for creating and maintaining database organization and integrity. They reorganize databases, partition files as well as install new or updated object programs used by the application. They must keep the Guardian operating system updated and coordinate with the HP site analysts when new hardware components are required.

Communications

There are numerous methods of communicating to and from an HP server system. The communication methods and the security that is necessary to control unauthorized users from getting access to the enterprise application is a joint effort by the System Manager and the Security Manager.

Communication devices, protocols, and methods are not covered in this book.

Protection and Recovery

The HP NonStop server was designed to be fault-tolerant. Depending on the system and enterprise applications, certain functions can be designed to take advantage of the fault-tolerant features of the Guardian operating system. The level of fault-tolerance will ensure the protection of and the ability to recover the application.

RISK The ability to pick and choose which functions will run as fault-tolerant will control the risk for the enterprise application.

The HP NonStop server hardware and software provide maximum protection against most equipment failures, power failures, and some catastrophic failures. For the enterprise application, the organization should carefully plan how to protect the database and application software.

Methods of Protection and Recovery ^[*] :

Mirrored Disk Volumes

Backup and Restore

NonStop Process Pairs

Pathway Applications

NonStop TMF software

NonStop RDF software

Mirrored disk volumes

Mirrored disk volumes are a primary protection against disk failures. Fault-tolerant mirrored disk volumes ensure that a single disk failure will not suspend the enterprise application. These volumes also provide the ability to repair and maintain disk volumes online, without interrupting application processing.

BACKUP and RESTORE

The BACKUP utility can copy one or more files or entire volumes to tape. BACKUP is used for database files and SQL programs stored in the Guardian file space. The RESTORE utility is used to retrieve data from the backup tapes.

NonStop Process Pairs

Applications that are written as non-stop process pairs, with application code to checkpoint the running process to its backup, give powerful protection to the enterprise application. If the CPU running the primary copy of the application process fails, the backup copy in the backup CPU takes over and the enterprise application is not affected.

Pathway Applications

Pathway uses the NonStop process pair concept and also uses an automatic load balancing and recovery mechanism. Although Pathway does not run server processes as non-stop pairs, Pathway automatically restarts a server if it fails. Pathway can run dynamic numbers of servers in multiple CPUs, allowing for load balancing and also allowing for CPU failures.

NonStopTMF Subsystem

The NonStop TMF subsystem provides database integrity through transaction control. Transactions can be defined to encompass multiple database updates that must collectively be successful or the transaction is backed out. NonStop TMF ensures that a database is consistently updated.

Remote Duplicate Database Facility (RDF)

Disasters like fires, earthquakes, or flooding can severely damage computer hardware. Catastrophes rarely happen, but when they do, they can take down the entire computer system leaving the company no way to carry on its daily business operations. Even a system shutdown of just a few hours can result in a substantial business loss.

The NonStop RDF subsystem monitors NonStop TMF transaction activity on a designated primary system and applies duplicate updates to an identical copy of the database residing on a remotely located backup system in the same network. The backup database is continually updated by NonStop RDF software, providing a current, online copy of the primary database. The backup database can be located nearby or across the nation, thus, RDF can protect against area-wide or even regional disasters.

^[*] These subsystems are described in more detail in the Gazette part of this book