Internet Traffic Measurement: Trends and Impact to Campus Network

Abstract— University of Lampung (Unila) is an Institution of Higher Education located in Bandar Lampung. Since 2016, Unila has deployed Internet Access Management (IAM) to guarantee the healthiness of the campus network, as well as to enhance the effectiveness of the bandwidth usage. This study focused on internet traffic measurement, conducted in Unila’s campus network during February 1 until February 29, 2016. Overall, this study shows user behavior on their application. The trend data of monthly most popular URL Categories accessed by users was; 1st Computers & Technology with 30032328 hits or 39.1%, the 2nd was Search Engines & Portals with 14214611 hits or 18.5%. There were around 30-40 % of internet traffic was use for Streaming Media activity, it proves that the existence of Streaming Media Activity in Campus Network which contribute to network congestion. During a month doing internet measurement, we identify the most active device/user that are the 1st was Aruba Wireless Controller with total traffic flow 40.45%, the 2nd was CCR-1 with 26.2%, the 3rd was CCR-2 with 16.9%, and the 4th was Digital Library Server with total flow was 0.6%. Monthly uplink traffic total flow was 5889.92 GB while downlink traffic total flow was 61041.35 GB. We made a recommendation to Unila management for implementing traffic provisioning especially on streaming media activity specific on access to Google Global Cache (GGC), to overcome network congestion during peak time period on working hours. Keywords—internet access management, internet traffic measurement; traffic tren