Silicom Denmark Session Based GTP Distribution
– Get control of your Load Balancing
The correct and on time distribution of GTP sessions is critical to achieving high performance analysis of the content within the sessions. The mobile operator’s BSS/OSS systems need to receive every correlated packet of a user’s or application’s GTP session within a specific timeframe or the usefulness of these expensive backend systems are severely reduced and their ROI is significantly crushed. Also impacted is the mobile operator’s ability to track subscriber activities, resolve subscriber issues promptly and be proactively notified of issues with critical revenue generating services.
Most application architectures rely on ‘all’ GTP signalling and ‘all’ GTP user plane being analyzed together in the same analysis engine. This ‘all traffic to the same engine’ requirement poses a significant problem in scaling the performance of such analysis engines. The scaling problems of this architecture, processing ‘all’ signalling and ‘all’ userplane together, can be alleviated by the use of external GTP distribution boxes distributing traffic via various keys such as IMSI, IMEI or randomly generate per-session keys etc.
Another architecture used is to process signalling and user plane separately and do on-the fly correlation. This architecture suffers from scaling problems due to the inter-process communication required. This stems primarily from a need for process synchronization of the engines’, in processing steps where it has a considerable impact on performance. Additionally, the inter-process communication and synchronization additional latency in processing. These shortcomings are not acceptable to most, and cost of hardware resources needed should be considered carefully.
This architecture can be assisted through devices offering inner IP distribution and signaling via one interface. But it does not offer real scalability and the shortcomings remain.
At the core of Silicom Denmark’s Session Based GTP Distribution are hardware integrated network traffic decoders and a GTP tracking engine. All based on high performance hardware, harnessing the power and flexibility of FPGAs.
The Silicom Denmark Session Based GTP Distribution is a distribution engine based on the Silicom Denmark hardware capture platform. As such, it provides the session and context management function as well as the fbCAPTURE Framework.
Dual level GTP session distribution
Silicom Denmark Session Based GTP Distribution offers a unique distribution mechanism that not only distributes traffic to interfaces, but also seamlessly provides distribution keys with each packet. This key enables the analyzing machine to also perform a second level of distribution to analysis engines. This allows for optimal utilization of the hardware platforms employed. Through Dual level GTP session distribution an unprecedented effectiveness and scalability can be achieved with new or existing analysis engines.
For maximum scalability and analysis a single node using Session Based GTP Distribution can forward channels to up to eight other analyzing nodes. This example is based on using fbC8XG Capture Card for GTP Session Based Distribution in a server. The fbC8XG has 8 x 10GE Rx interfaces used to acquire GTP traffic from the network. This distributing node acts as a hub in a cluster of analyzing hosts, probes or other devices. While at the same time it can perform traffic analysis by itself.
Using fbC8XG capture card for Session Based GTP Distribution it applies the power of nine servers or devices to the analysis of the GTP interfaces. While at the same time being transparent to the analysis engines themselves, in that no major architectural changes to the GTP analysis engines are needed.
The shift of data traffic from fixed line to mobile networks and the massive increases in mobile services now and in the future makes visibility and analysis crucial components for successful troubleshooting of subscriber issues and proactive notification of degrading application performance.
By utilizing the power and flexibility of FPGAs and the fbCAPTURE Framework Silicom Denmark’s GTP Distribution the capacity of analysis systems can be raised to a new level using existing implementations and devices. Hereby ensuring that monitoring systems can keep up with the growth in traffic volumes while providing same full and rich analysis as before. This can be realized on standard commodity hardware for analysis rather than ever more expensive hardware needed if trying to simply scale via pure processing power.