The redistribution of traffic within existing communication channels for budget optimization. To move from financing the expansion of existing channels to the construction of new points of presence.


    To optimize the throughput of the backbone network while constantly improving the quality of subscriber services. To implement monitoring tools, forecast network load and analyze the structure of data being transmitted so as to take preventive measures for modernizing communication channels that cannot cope with the load.


  • Deep Packet Inspection (DPI) technology operating at layers 2-7 of the OSI standard model
  • PacketLogic DPI System by Procera Networks
  • 3GPP (3GPP Policy and Charging Control)
Solve a similar problem


The preliminary stage was to complete four rounds of joint testing of the DPI solution under laboratory and real conditions on "live" traffic, including networks with a 10G bandwidth. The data obtained formed the basis for the principles of bandwidth allocation for all future nodes.

For the upgrade, we selected PacketLogic DPI systems by Procera Networks, specializing in the development of equipment for telecom operators and providers. The order for systems implementation was determined by the customer - the installation was carried out in stages, starting with the most heavily loaded channels.

Jet Infosystems engineers remotely checked and configured the equipment for each of the nodes (software upgrades, uploading traffic recognition signatures, etc.). In accordance with the implementation plan, ready-prepared devices with switching circuits were delivered to various cities and installed by VimpelCom specialists. The result was DPI hardware and software systems installed on 40 nodes at the borders of urban and backbone networks.

Upon completing installation, we enabled statistics collection at the nodes; this makes it possible to adjust the traffic control rules for each specific city.

Read more


DPI-based solutions allow traffic prioritization by allocating the bandwidth between different types of applications and services. After connecting backbone channels to new hardware and software systems and enabling collection of detailed statistics on the structure of user content, traffic management rules were determined for all regional sites. The customer is now equipped with a powerful tool for analysis and prediction of the node load, enabling proactive measures for the mitigation of the negative effects from traffic surges and the prevention of decrease in access speed.
Therefore, based on the statistics collected (with 60-65% of the load on the nodes coming from peer-to-peer applications) during peak hours, the priority of P2P traffic is reduced in relation to other content. During this period, the download speed from torrent trackers is slightly suppressed, while other services such as web browsing, online video, and messengers operates without delays. In addition, the operator can cache the multimedia content most demanded by users on local servers, thereby unloading backbone channels.
By setting restrictions for certain data types and significantly relieving the network, the company was able to redirect funds originally allocated for expanding existing channels to the arrangement of new points in the regions where it operations. This led to budget optimization and significantly accelerated development.
Thanks to 3GPP-based policy and tariff management architecture (Policy and Charging Control), PacketLogic DPI systems allowed the operator to provide personalized services - VimpelCom could then offer its customers tariff plans based on their individual needs. The company has thus significantly increased user loyalty and its competitiveness in the telecommunications market.
  • 60-65%

    Channel load comes from P2P applications

  • 1 year

    Time taken to install DPI devices in all operator cities

  • 40 nodes

    Upgraded with new hardware and software systems

All client projects

view more projects


I have read and agree to the privacy policy

By continuing to use this site, you consent to the processing of your personal data using the Internet services Yandex Metrika and Google Analytics. The procedure for processing your personal data, as well as the implemented requirements for their protection, are set out in the PDN Processing Policy. If you do not agree with the processing of your personal data, you can disable the storage of cookies in your browser settings.

Read more