Availability and Operability Metrics
Network item availability and operability help to answer the following:
- If the item is working at all, i.e. is it available? This aspect is reflected by Availability metric.
- If the item is fully functional, i.e. does it operate as expected? This aspect is covered by Operability metric.
 | Availability metric characterizes basic operation status of a monitored item. |
| Operability metric describes the ability of a monitored item to perform its essential functions. |
In Iotellect Network Manager availability is checked using simple standard procedures, like pinging a network device, or connecting to a service at specified port. The task is to ensure that a managed item is "alive" at all.
Operability monitoring is in contrast more complex and "in-depth" verification ensuring the managed item not "just works", but "works properly". Since the meaning of the term "works properly" differs for different monitored items, operability is usually much more complicated and very often a customizable procedure, like executing a remote script, sending and receiving email messages, etc. In this respect operability monitoring can be characterized as a smart monitoring. Operability checking procedures often provides more then just status of a monitored item, but acquire more detailed data from device.
Though the difference between these two metrics can seem somewhat subtle, relative and even subjective, it is still important to distinct these two metrics. Operability can be viewed as an advanced availability metric. For example, availability is only defined for individual services, but not for a whole network device.
Availability/operability check procedure depends on the type of monitored items:
- The basic IP network device availability is based on ping and, optionally, on traceroute services. In other words, availability in this case is defined as IP host reachability: the device is considered available if network packets can be delivered to and from this device.
- If an IP network device provides other services its availability is calculated as conjunction (logical "AND" operation) of all enabled service availability/operability statuses. Therefore the basic availability mentioned earlier is just an instance of more general procedure described here.
- Availability of standard IP host services is actually superseded by operability check. By default most services are configured for simpler availability check, but user can configure them service for in-depth operability testing if needed. See Network Host device driver chapter and it subsections describing particular services for some more details.
 | For example, SSH service by default has an empty script and therefore performs only connection, authentication and disconnection operations. This can be considered a basic availability check procedure. User can customize service by specifying a script. Script execution provides rich capabilities for advanced analysis of monitored device and therefore can be classified as operability check. Note, that the operability check includes all the steps executed by basic availability procedure. Hence successful operability check automatically mean service is also available. |
| For example, operational state of a network interface can be obtained using |
- Availability/operability of some SNMP-compliant devices can be accomplished using SNMP.
| See Printer, Wireless Device, and/or VMWare monitoring sections as examples. |
- Monitoring, including availability and operability monitoring, of other device types can require to implement a custom device driver.
| SQL database device driver was introduced for relational databases. It provides both availability and operability monitoring for JDBC-compliant database management systems. |
Iotellect Network Manager provides both current availability and statistical availability calculations.
See also Availability Monitoring section.
Was this page helpful?