2 Architecture

The system consists of several modules (servers) and they have the following functionality:

·       Apache server: Apache based user interface engine, the display of charts is handled by a separately started Chart server, while some of the WEB pages are generated by the separately started WEB server, which also handles the REST API calls. Both of them must be located on the same server as the Apache WEB

·       Oracle database: Stores the configuration and measurement data. Other than for starting up, the measurement servers do not need the central Oracle server, therefore no measurement is lost even under an Oracle database shutdown

·       GUI ↔ Equipment communication: Relays configuration commands and queries from the administrative part of the graphical user interface to the equipment directly (e.g. using direct SNMP communication) or indirectly through one of the data collection servers that can be assigned to the equipment. The access can be controlled on a server group basis. Direct access is recommended if possible, since it is significantly faster

·       Other servers: Each of these servers could be run even on different machines.

o   Application manager: Monitors the execution of the other servers, handles the automatic failover functionality, performs the self-testing operations and zips/deletes the old log and other files. Optionally, a second instance of this manager can run on another computer.

o   Job processing: In case of template-based equipment creation, the user can select later processing (if for example the equipment is currently unavailable) by storing the request in a job queue. In these cases the Job queue server creates the requested equipment later

o   SQLLDR: Inserts the measurement data into the Oracle database

o   Report calculation: Based on the measurement data, current time and the site - equipment hierarchy, the report engine calculates aggregate values, create reports and stores them in the database

o   Scheduled reports: sends reports on a schedule

o   SOAP server: it provides a standard Webservice interface that allows both the configuration of PVSR and the query of measurement and alarm data, therefore facilitating the easy integration of PVSR with other applications

o   Threshold processing: It calculates the current threshold violations based on the measurements

o   Event receiver: It receives and processes the SNMP Traps and Syslog messages. Optionally, a second instance of this module can run on another computer. By default the module expects the messages on the usual SNMP Trap and Syslog ports, therefore it should be run with root privileges on Unix/Linux systems

o   Data migration and Data compression: Transfers the data between the normal and archive tables, and compresses the measurements based on the measurement configuration after some time in one-hour intervals

o   Automatic template reapply: It is responsible for automatic measurement creation and deletion on existing equipments with the help of equipment templates

·       Measurement servers: they perform/collect the measurements based on the configurations in the database. In case of multiple servers of an identical type (e.g. Unix/Linux) the measurement servers can form a measurement server group (not shown in the figure), that supports automatic load-balancing with failover functionality. Based on its type, each measurement server makes its measurements either actively or passively, and either tracks or ignores the changes of the measured equipment.

Each module of the system (with the exception of the Chart and WEB servers) can be located on different machines, although it is recommended to run the SQLLDR, Threshold processing and Report calculation on the same machine.

PVSR can function in a fully duplicated/mirrored backup way. In this case every module, server; even the database is duplicated and PVSR does the synchronization periodically using an additional module called BACKUP. The PVSR backup solution does not require any additional third-party cluster service, like Oracle RAC or Linux cluster, etc…. The backup PVSR can be used in read-only mode all the time.

The application supports different operating systems and Oracle databases. For the complete list see the 13.1 section.

On the client side the following internet browsers are supported:

·       Mozilla Firefox version 3-

·       Google Chrome 4-