OSM supplies COSbatch to enterprises for batch job scheduling and application automation Security, management, automation, compliance - for UNIX, Linux and Windows

COSbatch architecture

COSbatch physical configuration diagram

The diagram on the right shows the way COSbatch components are distributed and connected.

COSbatch Server

COSbatch is typically implemented as a Server-Agent configuration, with all background and foreground processes scheduled from a single scheduling node running the COSbatch Server component. COSbatch Server must run on a UNIX or Linux system but may also schedule jobs on remote Windows systems. The COSbatch Server holds the repository of jobs and job suites plus definitions of the COSbatch objects – queues, processors, calendars, users, and user classes. One COSbatch Server is designated as the Scheduler Node.

It is possible, often desirable, to have multiple COSbatch Servers within a COSbatch domain. One or more COSbatch Servers may be designated as Control Nodes which synchronize their data repository with the Scheduler Node, taking over the scheduling in the event of planned or unplanned down-time of the Scheduler Node. This way, a COSbatch domain can be configured so that the Scheduler will automatically fail-over to a Control Node.

As the COSbatch Server software is network aware, multiple additional Servers can be run in parallel at larger sites, each being regularly updated with the information held on the others. When configured in this manner, the COSbatch Servers are designated as Standard Nodes. They are neither the Scheduler nor the Control (fail-over) nodes, but do have a synchronized copy of the COSbatch repository. An administrator can log into any COSbatch Server and view all jobs being processed within the entire enterprise. This provides a distributed point of control for administrators and help desk personnel, so providing a quicker mean time to resolve problems.

The benefit of being able to log in across multiple hosts must be traded off against the network traffic generated by multiple databases being continually synchronized. For this reason, the default configuration is to use one Scheduler Node and one Control Node, and use the lightweight COSbatch Agent to run jobs on the remaining systems within the domain.

COSbatch Agent

Any UNIX, Linux or Windows application server on which COSbatch is intended to control the running of background and/or foreground tasks normally runs the COSbatch Agent, a lightweight software component that handles all the instructions from the Scheduler Node for the system on which it resides. The Agent includes a secure communications method which can be changed to align with the organization's standards, for example by using SSH.

learn more about COSbatch architecture

 

Get your own trial copy of COSbatch COSbatch links Other OSM products

Privacy

Legal

Contact

Copyright© 2007  Open Systems Management Limited