Daemon Component

The Adabas System Coordinator daemon SYSCO is used by most installations, especially those using clustered applications, databases, or operating systems such as the IBM sysplex environment.

A daemon is required if you use any of these products or features:

  • Adabas Fastpath

    The daemon houses the Adabas Fastpath Asynchronous Buffer Manager.

  • Adabas Transaction Manager

    The daemon houses the Transaction Manager.

  • Single-system clustered applications such as UTM, CICS/MRO, IMS

    The daemon enables clients to be dynamically routed from one job to another with at rest session-related information being maintained by the daemon local to the client. Refer to Single-system dynamic transaction routing for more information.

  • Multi-system clustered applications such as CICSPlex

    The daemon enables clients to be dynamically routed from one system to another with at rest session-related information being maintained by the daemon local to the current instance of the client. Refer to Multi-system dynamic transaction routing for more information.

  • Single-seat current activity displays

    You can define your jobs to "pulse" activity statistics to daemon shared memory every so many commands or seconds, allowing information about any such job to be displayed in the Current Activity Displays option of the SYSCOR Natural application. Refer to Single-seat current activity displays for more information.

  • SAF-secured on-line administration

    The daemon houses the Adabas SAF Security component that allows a SAF-compliant security system to secure use of the SYSCOR, SYSAAF, SYSAFP, SYSATM and SYSAVI Natural applications

Daemon groups

You define a daemon group with a daemon job for each operating system image. The daemons communicate with other daemons in the group via XCF or Entire Net-Work, thus enabling multi-system dynamic transaction routing, cross-system Fastpath cache coherence, and cross-system transaction integrity. XCF is the recommended communication mechanism, if it is available.

There is normally a single daemon group with a single daemon running on each operating system image. Exceptions to this are:

  • Different workloads (for example, production and test) on the same system

    Again the daemons must be in separate groups and there must be a clear distinction between both client jobs and databases; that is, a test client job only accesses databases in the same workload grouping and databases are only accessed by clients in the same workload grouping.