Table of Contents
In a production environment, it is recommend that the TIBCO ActiveMatrix® Administrator server is hosted independently from TIBCO ActiveMatrix® BPM, by using its own set of TIBCO Host instances. If a high availability, fault tolerant configuration is required, the TIBCO ActiveMatrix® Administrator server should also be replicated. Finally the use of a distributed configuration for TIBCO ActiveMatrix BPM (along with a suitable underlying architecture) is recommented.
This configuration can provide the following advntages:
- Scalability :: TIBCO ActiveMatrix BPM software provides specialization and horizontal scalability capabilities. You can:
- add TIBCO ActiveMatrix BPM logical nodes to boost the capacity of the TIBCO ActiveMatrix BPM system in the required areas - e.g. process management or work management.
- distribute TIBCO ActiveMatrix BPM logical nodes to different TIBCO Host instances and physical machines as required.
- High availability and fault tolerance :: TIBCO ActiveMatrix BPM software provides active/active clustering capabilities. Adding a second TIBCO ActiveMatrix BPM logical node (of type BPM) provides high availability and fault tolerance. In the event of a system-affecting failure on one node, load is automatically switched to the remaining node.
Each discrete BPM system is hosted on its own set of TIBCO Host instances so they can be independently lifecycled or upgraded at the host level.
During the Implementation Phase of the first Project the following Areas should be considered and defined, before the first Go-Live. Following Projects can benefit from this decisions and already in place available techniques.
- Environment Configuration
- JVM settings
- JDBC Connections
- Resource Template Config (HTTP Client, HTTP Connector, LDAP connections)
- Database Tuning
- Scripted Deployment/DevOps etc.
- Deploy Scripts (install, and upgrade)
- Data Management/Growth
- Event Collector
- Timewindow to keep the data (data purging mechanism configured?)
- possibly use of an offline archive (cassandra?)
- Log Files
- Proper config, avoid excessive log statements
- appropriate log level
- Case Data Archiving/Cleanup
- Event Collector
- System Monitoring
- use of Tools like TIBCO Hawk, RTView, or others
- Monitoring of processes, disk space, cpu, JMX, etc.
- Backup/Restore, plan ensuring Data is in sync for recovery
- Disaster Recovery approach
- Is it required? Which approach (e.g. shared FS/in sync FS, or EMS queues in DB)
- Node Architecture
- highavailability/fault-tolerant, local network, Database setup?, EMS setup?
- Database configuration
- MS SQL, Oracle in highavailability/fault-tolerant Mode?
- Database Specialist must be part of the Operations Team (especially in case of Oracle RAC)
- Database should be configured with 'Auto-Extend'
- Production Machine Specification
- Machine Sizing e.g. enough memory, CPU, Diskspace, etc
- execute a proper performance Testing
- Load Balancer Configuration
- Configured with sticky sessions
- To avoid Cross-Site-Scripting Issues all parts of the BPM must be available with the same Domain Name and Port
- LDAP Directory
- in highavailability/fault-tolerant Mode
TIBCO ActiveMatrix BPM Applicitions are not aware of Nodes running in High Availability Mode and everything reconnect in case of an EMS or DB failure automatically.