Traffic management to improve network performance


10 February 2015 | Jet Info, Speaker: Daniil Vinyar, Alexander Gusev


Mobile operators’ revenue patterns are changing. Times when the subscriber base was growing at a healthy pace to generate higher revenues and the average revenue per user (ARPU) was on the rise as well are largely a thing of the past. Operators now face a stagnating market where network traffic consumption keeps increasing (indeed, an average household that used to have a single computer with an Internet connection now has several wi-fi devices) but the ARPU tends to fall because of high competition. To meet the growing consumer demand for data, carriers and internet providers are forced to increase capital expenditure (on the expansion of existing networks) as well as operational expenditure (on running additional hardware and channels).

According to the Russian Ministry of Communications, mobile phones were used by 164%, mobile Internet by 55% and landline Internet by 46% of the population in the country. According to Internet World Stats, Russia has the largest number of Internet users in Europe (Fig. 1).

Fig. 1. Number of Internet users in Europe by country (Source: Internet World Stats)
Number of Internet users in Europe by country (Source: Internet World Stats)

Forecasts indicate that a situation may arise when network operating costs would outstrip revenue (Fig. 2). To mitigate this risk, the operator should focus on retaining its current customers and slashing costs rather than expanding the customer base. An operator’s core business is to deliver content from its source to the user. In other words, the transmission cost per megabyte should be lower than what the customer pays for this megabyte.

Forecasts indicate that a situation may arise when network operating costs would outstrip revenue. To mitigate this risk, the operator should focus on retaining its current customers and slashing costs rather than expanding the customer base. An operator’s core business is to deliver content from its source to the user. In other words, the transmission cost per megabyte should be lower than what the customer pays for this megabyte

Fig. 2. Forecast of network costs and operator revenue against the background of increasing network traffic
Forecast of network costs and operator revenue against the background of increasing network traffic

Fig. 3. Estimated costs and revenues per megabyte delivered to the customer
Estimated costs and revenues per megabyte delivered to the customer

The cost of delivery per megabyte depends on the distance between the data source and the user. The distance in this case is defined as the number of nodes and communication channels that are involved in transferring data. Accordingly, the longer the distance, the higher the cost of delivery. Let us analyze this cost in the case of a typical large operator (Fig. 4).

Fig. 4. Sample topology of a major telecom and wireless operator (own backbone, landline and wireless services)
Sample topology of a major telecom and wireless operator (own backbone, landline and wireless services)

Data transfer within a single access segment uses a minimum of hardware and entails the lowest cost. Delivery to the neighboring access segment is somewhat more expensive. External national peering would cost still more. The most expensive operation would be to deliver data abroad (external international peering). Note also that delivery through the mobile network would be more expensive that through the FTTB (fiber to the building) network.

Mobile operators’ revenue patterns are changing. Times when the subscriber base was growing at a healthy pace to generate higher revenues and the average revenue per user (ARPU) was on the rise as well are largely a thing of the past. Operators now face a stagnating market where network traffic consumption keeps increasing (indeed, an average household that used to have a single computer with an Internet connection now has several wi-fi devices) but the ARPU tends to fall because of high competition

Moreover, to ensure customer loyalty the operator should realize their preferences and understand the value of content they want. Subscribers essentially use two groups of applications, namely bandwidth-sensitive (peer-to-peer, video streaming, file exchange and so on) and delay-sensitive ones (interactive applications such as online games or Skype). In view of the above, the operator needs a tool for traffic analysis and management. Such tools are called DPI (deep packet inspection). A good return on investment requires full bandwidth utilization (by prioritizing network traffic, Fig. 5) while maintaining customer satisfaction. DPI systems establish traffic priority according to certain rules (Fig. 6).

Fig. 5. Example of controlled congestion
Example of controlled congestion

Fig. 6. Example of traffic prioritization
Example of traffic prioritization

Additional tools are available for reducing data transfer costs and increasing the value of traffic for the user to ultimately generate additional revenue. They are described in detail elsewhere in this issue.

More information: http://jet.msk.su/en/about/


back

READY TO TALK ABOUT YOUR NEXT PROJECT?

×

Hello!

See examples of Jet Toolbar messages!

Find the icon in the bottom right-hand corner of this page.

Click on it and see some examples of messages that can be created with Jet Toolbar