Network traffic monitoring and analysis is of theoretical and practical significance for optimizing network resource and improving the user experience. However, existing solutions, which usually rely on a high-performance server with large storage capacity, are not scalable for detailed analysis of a large volume of traffic data. In this, we present a traffic monitoring and analysis system for large-scale networks based on Time data. This system has been deployed in the core network of a large cellular network and extensively evaluated. The results demonstrate that the system can efficiently process 4.2 Tbytes of traffic data from 123 Gb/s links with high performance and low cost every day.
Estimation based execution assessment of system movement is essential for the provisioning of oversaw and controlled administrations in short timescales, and additionally for empowering the responsibility of system assets. The enduring presentation and organization of the Internet Protocol Next Generation (IPNG-IPv6) guarantee a system address space that can oblige any gadget equipped for creating a computerized heart-beat. Under such an omnipresent condition, Internet movement estimation happens to specific significance, particularly for the guaranteed provisioning of separated levels of administration quality to the diverse application streams.
The non-indistinguishable reaction of streams to the distinctive kinds of organizing forced execution debasement and the predictable development of arranged gadgets raise the requirement for pervasive estimation systems that can be similarly pertinent to various applications and transports. This theory presents another estimation method that endeavors local highlights of IPv6 to turn into a basic piece of the Internet’s task and to give natural help to execution estimations at the all-around display organize layer. IPv6 Extension Headers have been utilized to convey both the triggers that summon the estimation movement and the momentary estimation pointers in accordance with the payload information itself, giving a high level of certainty that the conduct of the genuine client activity streams is watched.
The in-line estimations instrument has been fundamentally looked into to existing estimation procedures, and its outline and a product based model usage have been reported. The created framework has been utilized to temporarily assess various execution properties of an assorted arrangement of utilization streams, over various limit IPv6 test setups. Through experimentation and hypothetical argumentation, it has been demonstrated that IPv6-based, in-line estimations can shape the reason for exact and low-overhead execution evaluation of system movement streams in brief time-scales, by being powerfully sent where and when required in a multi-benefit Internet condition.
Processor: Pentium-IV 2.6GHz
Hard Memory: 40GB
Front End: ASP.Net
Back End: Microsoft SQL Server 2000
Working System: Windows XP
Framework: Microsoft Visual Studio 2005