Research
Measurement and Classification of Network Traffic Analysis Using Hardoop
Network traffic can classified as a process which list computer network based on some parameters like port number and protocols into some traffic classes like undesired, sensitivity etc. Traffic can be implemented differently to differentiate the service required for the user for the specific purpose. The large demand of increase in internet users and increase in bandwidth required for various applications are escalating day by day. The traffic data needs to be classified and analyzed with certain tools. Hardoop is the tool which performs the task in very time efficient manner. Hardoop actually run on commodity hardware which process this huge data with hive. Traffic analysis, measurement and classification are done by hardoop based tools at various parameters of packet and flow level. The derived result is used by network administrator for resolving networking related issue. The measurement of internet traffic and analysis has been implemented from long before but the problem is recent years the user in internet has escalated dramatically. We proposed network traffic management system for analyzing internet traffic of multi-terabytes in extensible manner to perform HTTP, ICMP, UDP, TCP and IP.
Study of Effective Scheduling Algorithm for Application of Big Data
In this new era with the advancement in the technological world the data storage, analysis becomes a major problem. Although the availability of different data storage component like electronic storage such as hard drive or virtual storage such as cloud still the problems remains. The major issue is processing the data because usually the data is in several format and size. Usually processing such huge amount of data with several formats can be time consuming. Using of application such as Hadoop can be beneficial but using of scheduling algorithm can be the best way to for data set analysis to make the process time efficient and analysis the requirement of different scheduling algorithm for the specific data set. In this paper we analysis different data set to explain the most effective scheduling algorithm for that specific data set and then store and execute data set after processing.
