Performance Manager

Executive summary:

Any performance monitoring tool requires the extraction of performance measurements targeted at a wide range of specific types of data. REWSStools® extracts numerous measurements and statistics directly from the network elements themselves for the purpose of quantifying the performance of the network across a broad spectrum:

• NSS Performance Stats
• BSS Performance Stats
• Calling and Mobility Performance Stats
• Traffic Performance Stats
• Load Performance Stats

These measurements are useful assets in determining whether specific components of the network are functioning as expected or if their substandard counter values are the indicating problems on the network.

Many types of measurement data are collected at regular intervals for all types of network elements. The default collection interval is 1-hour but some intervals will vary. The goal is to collect with the finest level of granularity possible without unduly loading the system. Once collected, REWSStools® provides numerous interfaces for viewing and interpreting the data.

Concise summarized reports can be dynamically generated across any time frame including month, week, day and hour. Dynamic charts make it possible to view the activity of any counter as it is trended over time. High level reports and charts are linked to provide simple drill down features.

Pulling Performance Data Directly from the Network Elements

There are multiple ways an application can use for collecting performance data from network elements. Each has its strengths and its weaknesses.

1. Signaling Probes: Using probes and/or protocol analyzers
   1. Enormously powerful but also enormously expensive
   2. Vast amounts of data but more attuned to detailed investigations rather than network monitoring
   3. Comprehensive data about links between NEs but not about the internal happenings within the NEs
2. CDR Analysis: post-processing of CDRs
   1. Valuable source of data in all successful calls, but not useful for failed calls.
   2. Post processing approach meaning the real time nature is compromised
3. Switch Data: Using the counters, states, & alarms generated by the elements themselves
1. Minimal load since all data is processed internally on the switch, although data is not as granular as with other methods

REWSStools® has been designed to extract its data directly from the network elements themselves since this is the best approach to understand what is happening inside the switch and associated elements. Any problems found here are usually clear indications that customer services are being impacted.

The value for the operator:

Enterprise-wide visibility of all pertinent performance measurements and counters is imperative for detecting and diagnosing negative trends or sudden changes in network performance. Being able to target any performance criteria and access detailed counter values from finely granular intervals is a very powerful analytic opportunity. Gaining perspective over this type of data in a near real time environment is critical in preventing customer affecting outages.