Net Neutrality

Does lack of Net neutrality destroy the benefits of network monitoring?

Tuesday, April 10, 2012 | Julian Palmer

What is Net Neutrality? In simple terms it is the principle that all Internet users are equal and there are no restrictions imposed by Internet service providers on user's content, on the publisher or the content, on the usage of the content or performance of the flow of content.

I have often used the similarities of road traffic to provide a more human example of the challenges that affect time dependent applications that use the public network. Road traffic enjoys frequent delays caused by the same threats as network traffic – namely congestion & regulation. These threats become more invasive when ‘time' is a material element of the trip. On a road trip that means getting to a destination on time; on the Internet it's the equivalent of having the application you're using working properly – for example VoIP call voice quality so poor that you give up half way through the call; or your ecommerce purchase to get the new iPad3 on its release day not failing because of the massive demand for the product.

[ White Paper - Understanding Internet Speed Test Results - The problem is not in the measurement, it is in understanding the test results as they relate to the application problem being experienced (PDF) ]

But returning to the road traffic analogy, what if roads were not neutral to all its users? Imagine for a moment that the big car manufacturers built and owned many of the highways and imposed their management credo on those that wanted to use the highways they owned. It is entirely likely for example, that the Ford manufacturer might provide a priority lane on its roads for Ford cars at the expense of other manufacturers. Why would they do this? Money of course. The simple principle of favouring those customers that contribute financially is the basic principle of business survival and growth.

So how does this simple change of neutrality on the road alter how we monitor and manage traffic for service and performance? After all, this is the prime objective of monitoring data networks.

If the roads were not neutral it would change the dynamics of how we select cars and it would also dramatically change how we measure the performance and service of the roads we use. For a start we would be inclined to validate road coverage and performance by car manufacturer just as we do with mobile phone providers today. We would want to choose a car that delivered the best service for our everyday needs like the school run or the commute to work. If the main road that you commute into the city every day was a ‘Ford road' you would likely choose a Ford car and would see advertisements like ‘New York just 10 minutes from JFK for Ford drivers' to encourage this choice. Can you imagine the politics and the chaos that would result? Could or would a car manufacturer block entire access to a road for a competing manufacturer? Could a car manufacturer actually own all the major roads into a single city? How would small car manufacturers compete against the large wealthy manufacturers?

For data networks the problem is more severe because the traffic is not visible to the naked eye as on the roads. Any network or application provider can monitor networks but can the principles of monitoring address performance issues that are the result of regulatory policies? It was reported on  that a customer of AT&T sued and won because AT&T throttled his connection of an unlimited data plan. According to the publication, the customer only became aware of the restrictions because AT&T communicated their policy to him directly when advising him that his usage was too heavy. The vital question is what would have happened if AT&T simply throttled the connection but did not tell the user. By slowing traffic, applications that are dependent on timely delivery such as media applications will start to fail. If your business had been delivering media applications, you would most likely have received a support ticket from your customer complaining of poor media quality. However, your network monitoring would almost certainly report that your network service was healthy. Could you lose a customer because of the AT&T traffic policy? Absolutely.

[ White Paper - Understanding Internet Speed Test Results - The problem is not in the measurement, it is in understanding the test results as they relate to the application problem being experienced (PDF) ]

As another example, in 2008 a business customer reported that application disconnects were causing a number of problems for the execution of their day to day business while their wealth of network monitoring data revealed no material issues.

Implementing a focused testing solution quickly revealed that the consistency throughput of TCP packets was erratic, which delivered low throughput, while HTTP packets (also TCP) were consistently delivering a high throughput. When the company raised this issue of protocol level discrepancy, the service provider revealed that it was its policy to have all HTTP traffic take priority over non HTTP traffic. This regulatory policy clearly contradicts the definition of neutrality and was enforced solely for commercial reasons because HTTP traffic represented the majority of the service providers business. This ‘neutrality' threat resulted in the customer switching provider.

So if Net Neutrality is discarded to support commercial interests, does Net Neutrality exist at all?

Monitoring validates availability, usage and consumption with the purpose being to provide an assessment of potential service level risks. In network terms this assessment determines if the road is open and flowing freely such that there are no ‘determined' unnatural delays. The problem of course is that any provider defined policy impacts the flow of data by choice and thus, by definition, does not constitute an unnatural delay. The definition of ‘freely' does not mean that the performance for the customer is acceptable. Essentially in the two examples above, the monitoring of the network may reflect that it is operating as intended because it is conforming to policy but that policy may not be acceptable to the customer.

It should also be understood that not all applications are equal. The importance of a customer application and its demand on the network is information that a service provider is not privy to when determining policies that affect the flow of data through the network. If a network owner does not publish the fact that regulatory policies exist, how does a business or a consumer know that the application problems experienced are not intentionally manufactured? This is a crucial question because, unlike road traffic, network policies that affect the neutrality of the network to benefit the network operator are not visible. Systematic active testing, not monitoring, provides the solution framework to address this issue.

About the Author

Julian Palmer has spent more than 30 years in technology specialising in operation management and service delivery. In his current role of Chief Technology Officer of Visualware, Julian has leveraged his experience and detailed knowledge of application data flow to evolve solutions for measuring and reporting network quality.

Founded in January of 2001, Visualware, Inc. is a leading creator of solutions to manage and report the true customer and end-user Web experience.

Visualware’s Enterprise solutions include a suite of products and OEM technologies that enable any business or organization to view, report and troubleshoot a customer's actual transaction response time anywhere in the world -- as it happens in real time.    Learn more about

  • Print
  • Send to a friend