PVSR 3.3
Reliable performance monitoring for all size of enterprises and service providers
The PerformanceVisor (PVSR) is a comprehensive performance monitoring system for monitoring, maintenance and reporting purposes. It can monitor different systems including servers, networks and applications in enterprise or service provider environments. PVSR is a useful tool for operators, making their daily tasks easier, but has other uses too: it can provide information for the users about the usage and load of the systems they are using and make the capacity planning easier by identifying the most used resources in a system. Thanks to its flexible and scalable architecture and to its license policy, PVSR is fitting for systems with various sizes, monitoring tens or thousands of equipments.
PerformanceVisor - following the current trends - offers with its reporting subsystem more than measurement summarization, represented in tables (normal or TopN-BottomN) or depicted on charts. With its complex threshold subsystem the users can define on the GUI even thresholds capable of representing a Service Level Agreement (SLA). The two subsystem make it possible to create SLA reports, showing the percentage of time when SLA violations had occurred. In some cases this isn't even enough, so PVSR has also an SLA subsystem extension which can pull data from other (for example trouble ticket) systems as well and compute more complex SLAs based on these values.
Main key features:
- Easy-to-use WEB based GUI: PerformanceVisor has the same WEB based GUI for administration and for normal daily use purposes. The user interface is using Ajax technology making it possible to have functionalities to which the users are used to with non-browser based applications. PVSR also makes the work of the administrators easier, by providing templates to the most common configuration tasks.
- Diverse, distributed measurement system: PerformanceVisor can run in a distributed environment using multiple server nodes. With this feature on the one hand the software doesn't require a big server for large monitored systems, and on the other hand the number of nodes running PVSR can scale with the growing of the monitored system. The system continuously checks its components and recognizes the critical events occurring during operation and immediately notifies the operators about them. PVSR has several purchasable off the shelf measurement types and a documented programming interface (API), and with its use even the customer can create new measurement types. Moreover the configuration of new SNMP measurements can be done on the GUI and so in that case doesn't require any programming skill.
- Customizable object hierarchies and charts: During the configuration every equipment and measurement are created in a hierarchy under one of the sites in PVSR. The users can immediately see and use this hierarchy but they can also create their own: grouping those equipments and/or measurements together which are important for them, creating new charts with different measurements on it or even modifying the menu structure of the application. The administrator users can share this alternative hierarchy with the other users, creating parallel hierarchies in one PVSR.
- Complex threshold subsystem: PerformanceVisor is capable of evaluating complex expressions, making it possible for the users to create threshold definitions which accurately represent error conditions or SLA violations. During evaluation the measured values can be compared with static and variable values; in latter case the system automatically calculates these based on the statistical data (mean, deviation) of the past measures. A threshold can be based on several measurements and they can be bulk-created using templates as well. The system has a group threshold capability with which it can raise a new alert if the given number and type of alarms is present at a specific point in the hierarchy. The thresholds can be organized in hierarchy: if a higher level alarm is raised then it masks out the ones at the lower levels, thus creating a root cause like analysis. The alarms can be viewed on an on-line alarm window, on measurement charts and on violation charts (showing the number of violations at any given time). The users can acknowledge the violations and the system stores the acknowledge time and user id as well. PVSR can notify other systems about these violations.
- Reveivig alarms: PVSR is capable receiving alarms from 3rd party systems or directly from monitored equipment in the form of SNMP Trap and syslog messages. PVSR conciliates the received alarms with its own collected measurements, issues alarms and visualizes them with predefined threshold alarms.
- Report subsystem: Summary reports can be defined on arbitrary segments of the hierarchy. PVSR has tabular report views (which can be reordered by the user at any time) and chart report views (for example: showing the sum of interface traffic on an equipment at any time). The threshold violations can be summarized in report views as well, showing the percentage of time when a violation occurred, allowing to create simplified SLA reports. The system is capable to send these reports regularly via e-mail on a daily-weekly-monthly-yearly basis.
- Configuration: The configuration of PVSR isn't just aided with an easy-to-use and ergonomic user interface, but with templates and a northbound SOAP interface as well. The templates make the creation of equipments, thresholds and charts even more easier to the users and the SOAP interface provides a standard based API through which both the query of the measured values and the configuration of the whole system is achievable, thus by using it PVSR can be easily connected to other systems (for example to an inventory system).
Other key features:
- Every measured or calculated data (for example reports and violations) can be exported into Excel files
- Fine grained security system
- Configurable schematic maps for representing objects and relations between them. For example: a map showing the sites and the communication link between them. The elements on a map can be linked to the PVSR objects of course
- Audit trail system, showing the changes in the configuration along with the modifying timestamp and user name
- Time windows for the report and threshold subsystems allowing to suspend the calculation/evaluation. For example: violation can occur only weekdays during working hours
New features in PVSR 3.3
Baseline thresholds
Receiving SNMP Trap and syslog
System self tests
More effective configuration
Simplified navigation
Integration with any external WEB interface
Measurement of Cisco CallManager calls
References:
Alcoa
Antenna Hungaria
Elektro Ljubljana
Invitel
PublicNet subnet NOC
PublicNet NOC
Molari
National Healthcare
Pantel
T-Com
T-Mobile
Measurement server types:
SNMP
Netflow
ASCII
Oracle
Ping
Synthetic transactions (HTTP, FTP, DNS, DHCP, NTP, RADIUS, TELNET, SSH, Email, LDAP)
Cisco QoS
Unix/Linux
CiscoPING
Cisco SAA
Cisco CallManager