QoS: Solution Waiting For A Problem

Quality of Service (QoS) in data networks has been debated for atleast 35 years. Despite two decades of vigorous research and standards activity, QoS schemes have floundered. Research in QoS mechanisms assumes the existence of a large market for QoS-sensitive applications that will justify the expense of enabling QoS in the network. However, the lack of need for hard guarantees by both the customer and the internet service provider, is one of the fundamental reasons for the failure of QoS mechanisms. QoS research seems to provide a solution to a non-existent problem that may never arise. In this position paper, we argue that it is needless to invest in building QoS solutions. The widely used overprovisioning technique is a sufficient and remarkable traffic engineering approach.