It was 15 years ago. VoIP (Voice over IP) or IPT (IP Telephony) was not well developed or common. I was an engineer at a global Telco company, responsible for service enablement. At that moment, some of global companies were doing heavy expansion towards Asia - then-emerging markets like China. Effective communication became one of the critical factors for these global companies to run their business in different countries and timezones. Certainly, there are many ways to communicate internally and externally. This could be mail, phone calls, or fax, but no one would disagree that a phone call was the most simple and efficient mode of communication in 2000.
But when the IT managers got the IDD (International Direct Dialing) bill, the monthly cost on legacy voice infrastructure was huge and increasing every month. This situation was especially significant in the emerging markets because government authorized the Telco company to monopolize the market. IDD charges would remain very high until the government allowed competition. Many CTOs found it very difficult to justify the business value brought by these overly-expensive IDD calls. I still remembered helping one big logistics company calculate the ROI in implementing VoIP solution. The conclusion was they could save more than 40% of spending on the IDD every month if they migrated to the VoIP. Not surprisingly, every head of IT was eager to investigate whether the investment to move from legacy voice infrastructure to VoIP, using the converged data and voice service across their WAN links, was now worth it.
The popularity of VoIP really took off around that time in Asia. Companies first started to deploy the VoIP solution in a conservative fashion. At first, most companies decided not to fully replace their TDM PBX infrastructure into full IP based voice infrastructure. They preferred a hybrid mode in order to minimize the investment and test to see what the end-user experience would be like, and the general trend of cost-saving. The hybrid design established the physical connection between the TDM PBX systems with the VoIP enabled routers. The analog voice was converted and compressed to digital signal by the DSP (Digital Signal Processor) on the routers and then transmitted to the other end of the office via the WAN links.
During that time, IT teams found much more end-user complaints on voice quality after the migration to VoIP. Now, voice quality can be very subjective. Different people may have different level of acceptance in determining what is good voice quality. IT operations teams usually do not have any proactive VoIP monitoring solutions to measure performance in real time. Once an IT team received end-user complaints, the first thing they would do is to access the NMS to check the one-second average bandwidth utilization of the WAN link. To be honest, most of the time you would find the link utilization looked good until you directly accessed the routers. There you would find the queuing drop counter had increased significantly. This indirectly indicated that the WAN link had been congested and some voice packets might have been delayed or dropped, most likely because some bursty traffic consumed all the bandwidth in an extremely short period of time. But what was that burst traffic? In those days, I am afraid no one could tell. Besides, at this point we had not solved many of the compatible issues between TDM PBX and IP voice routers.
Nowadays, VoIP or IPT is very popular and very common. Just about every company has IP phones across most offices. Control protocols like RTCP (RTP Control Protocol) are able to provide out-of-band statistics to show the quality of voice calls (MoS), latency, packet loss, packet delay, and more. VoIP monitoring solutions should capture these control messages and extracting their contents from network data provide real-time statistics on the quality of each voice call. Real-time visibility on bursty traffic is a powerful tool for IT Operation teams in managing the converged voice and video data. This visibility switched the IT operation team from a reactive mode to a proactive mindset, making them able to monitor voice quality in real-time and have much more useful statistics in planning for any expansions on the network infrastructure. It is indeed a great improvement on how we used to monitor voice services 15 years ago.
So ask yourself: do you and your IT Ops team have full, real-time visibility into your network performance?