Prospects of Volunteer Computing Model in Performance Data Gathering for Broadband Policy Formulation: A Case Study from South Asia

Abstract

The recent unprecedented growth of telecom facilities has offered the Internet users in most Asian countries a flavour of broadband. Yet, despite rosy promises by telcos, the user experience has often been less than ideal. High cost of infrastructure, particularly of international links, prevents them from enjoying undisturbed access to internet. These challenges can only be overcome by right policy decisions based on evidences. Thus, monitoring the broadband Quality of Service Experience (QoSE) becomes more than an attempt to ensure quality delivery. It also creates a basis for policy formulation.The first approach to monitoring QoSE, is the regulator reaching deep into the innards of the telecom network to install monitoring equipment and taking remedial actions, specified under the licenses or the governing statute, when the data indicate below‐standard performance. Dearth of financial and human resources can be the key challenge in such a direct approach. The second approach is based largely on user activism. Educated users are expected to voluntarily contribute their time and computing resources towards building a performance database which in turn will be used in creating the bigger picture.A comprehensive methodology to benchmark Broadband Quality of Service Experience (QoSE), based on the latter approach has been developed jointly by LIRNEasia and TeNet group of Indian Institute of Technology (IIT) Madras. While there is no barrier for regulators to use it, the methodology is largely user centric. Instead of depending on one time pinging, this methodology uses AT‐Tester, an a open source based software tool to monitor all crucial QoSE broadband metrics over a longer period, on both week days and week days covering peak as well as off peak traffic. The traffic is also monitored within segments, ISP, local and international.The methodology adapts the concept of Volunteer Computing (or Public Service Computing), where complex computing tasks are broken up into small chunks and are then run in the background of large numbers of computers of volunteers who are simultaneously engaged in other tasks. AT‐Tester is installed in a large number of computers that are connected to the Internet and run in the background. The outcome is aggregated in real‐time on a server and made available through site www.broadbandasia.info. This approach would take the quality of the results to a whole different level, averaging out anomalies and allowing continuous coverage. The paper analyses how this approach could be used in broadband policy formulation taking examples from Bangladesh, India and Sri Lanka, the three countries where the experiments have been conducted.