With NextNine, CEO, Adi Dulberg

Q: What value do remote monitoring and proactive support offer IP Communication service providers that mainstream solutions do not?

A: Unlike mainstream, reactive methodologies, proactive, remote, support automation solutions perform scheduled preventive monitoring and maintenance that avert downtime and prevent problems from impacting service – even before customers identify initial problem symptoms. This approach ensures that resolution is initiated at an earlier stage in the problem cycle, resulting in lower mean time to repair (MTTR) and maximum system availability.

In addition, remote diagnostic tools are designed to monitor and analyze telephony networks, providing support engineers with a consistent, timely view of all problems identified by diagnostics routines in all devices/applications being monitored. The result: dramatically accelerated problem prevention and resolution, as well as maximum system uptime.


 
Q: With consumer demand for converged communications applications on the rise, and the complexity inherent in the underlying architecture, such as IP Multimedia Subsystem (IMS), what additional considerations must telcos take into account when considering remote monitoring and proactive support?

A: There are two primary issues that must be addressed: security and knowledge effectiveness. Converged communications clearly represent the future of telecommunications. As such, no matter what IP-based communications methodology you’re talking about, first and foremost on everyone’s mind must be security. All remote access traffic must be encrypted over the Internet, regardless of the type of access used. Several successful implementations of which I am aware use only port 443, SSL-based, outbound only communication, even for remote access sessions, file transfers and all other remote workflows, creating an ultra-secure, customer-accepted operating environment.

The complexity of converged communications solutions holds a major risk when it comes to monitoring, such as the inherent danger of flooding the monitoring organization with huge number of false alarms. Traditional approaches to monitoring are based on receiving all alarms and related data available from the system’s various units at the device level, then subsequently correlating this flow into meaningful alarms.

The problem is that with the increased complexity of these systems, correlation is almost impossible, rendering much ‘application level’ monitoring unusable.

A different approach is to start from the problem level, using post-deployment knowledge to learn what are the most critical problems, and how they can be avoided.

By monitoring for specific system-wide problems, the number of alarms generated can be reduced to the minimum level required to maintain high availability, without creating an enormous amount of ‘noise’. Also of note, is that this approach requires tools that use agent-less monitoring, and enable rapid deployment of new knowledge when it is manifested in the field.