Tracking your Application Time

Apr 10, 2015: Mobile devices have become an integral part of our lives, however, most of us have little idea how often and how long we are spending with these devices. Track your mobile phone habit by installing Application Time (AppT) today!

Mobile devices have become an integral part of many of our lives, so much so that recent studies have found severe separation anxiety between people and their mobile phones, with participants citing the apprehension of missed communication with friends or loved ones without access to chat or social media applications. Motivated by this as well as anecdotal evidence of the large effect of mobile devices on our lives, we at the Northwestern Aqualab are embarking on a study to collect and analyze the usage patterns and habits of the population. Today, we are releasing our mobile application to help users track, analyze and understand their relationship with their mobile device. Named Application Time – or AppT for short – the app allows users to track and visualize their usage of mobile applications and the frequency and duration of their mobile device interactions. The application is now released to the Android Store for download – a more detailed description of the measurements and mechanic, refer to the project page on our website (

The screen shots below show how the app allows users to visualize aggregated and individual usage of applications on their device through an easy, intuitive interface. The app also allows users to export their usage data so they can perform their own analytics if they desire.

We have been using AppT for over a year now internally in out lab. To illustrate some of the awareness and insights that this can bring, I present an analysis of my own mobile device usage. The first interesting thing we found was the long tail of application usage, and that the top 10 apps account for over 90% of the device time, and that the other 30 or so apps only account for the remaining 10%. These top apps (shown in Table 1) include common communication related tasks such as calls, SMS and email. It was surprising to see Google Hangouts – a SMS and chat client – was used over 11,000 times over the course of a year; that amounts to just over 23 times a day on average.

ApplicationInvocation Count
Google Hangouts11511
Chrome for Android1540
Transit Tracks (train tracker)1322
Google Maps820

Looking at temporal usage patterns, a couple of interesting items can be seen. In the figure below, each dot represents an instance when a mobile application was launched. Besides the overall amount consistent quantity of usage, we see that in general the phone is the first item checked every morning, and is consistently accessed throughout each day and into the early morning.

One point of concern was the peak in usage each day during the middle of the work day in the late afternoon. It was noticed that the device was primarily being used for chat and email messaging even though the user was sitting in front of a computer. This meant that work is being consistently interrupted to check each and every notification on the device, regardless of the notification’s urgency. Such a constant distraction can be very detrimental to long period of focused effort. Once aware of this activity, the user now disables all notifications throughout periods of concentrated work (using several of the apps available), and has seen great improvements in productivity.

Our lab has released the app Application Time to the public on the Google Play Store so others can collect and analyze their mobile device usage. Hopefully, by using AppT you will be able to figure out ways your own ways to break dependence, improve productivity, or be aware of your device habits.

Getting in and out of Cuba

Sep 9, 2015: Change is coming to Cuba - Our early work on characterizing Cuba's Internet connectivity to appear at the next ACM Internet Measurement Conference (IMC) 2015!

It may have taken 54 years, but change is coming to Cuba. Last December, the US government announced plans to restore relations with Cuba and ease restrictions on travel and trade. It took little time for American businesses to start scouting opportunities - from Netflix to Airbnb.

Despite the promising news, the state of Cuban infrastruc- ture, particularly in the computing and network segment, pose no small challenges to these plans. Today, less than 5% of the population have their own fixed-line Internet connection and only an estimated 25% of the population are able to get online. Those that are actually connected experience very poor performance. Ookla’s NetIndex, for instance, ranks Cuba among the worst ten countries in terms of average bandwidth – 197th out of 202 – with a measured broadband download speed of 1.67 Mbps.

We have started to characterize the state of Cuba’s access to the wider Internet. Our first paper on the topic - appearing at the ACM Internet Measurement Conference 2015 - reports on some of our early findings, including high RTTs to websites hosted off the island, even after the addition of ALBA- 1, a high degree of path asymmetry in traffic to/from the island that partially traverse high-latency satellite links, and several web services that return invalid responses to requests originating from the island. We plan to make a periodic status report on the state of the Internet in Cuba and the associated data available to the research community.

Mobile Network Performance and Content Delivery - Part 2

Mar 20, 2014: Most Content Delivery Networks (CDNs) attempt to pick the closest replcia server to your location. We see large inefficiencies in replica selection for cellular networks, with remote Public DNS resolvers retrieving better replicas a significant fraction of time.

Led by John Rula

This series looks at the current state of network performance and content distribution on cellular networks. Download Namehelp Mobile for Android in the Play Store to find which DNS give you the best performance on your phone. In Part 1 we looked at the impact DNS performance and compared your default DNS service on your mobile device to several public options. In this part we look at how Content Delivery Networks (CDNs) attempt to optimize content delivery on mobile networks.

Content delivery networks (CDNs) are the unseen engine which drive the Web. Companies like Akamai and Limelight operate a global network of replica caches, with the intention of delivering content from the closest (and thus fastest) server to its users. For instance, when you go to the New York Time's website, your browser resolves to its IP address by its locally configured DNS, typically provided by your internet service provider. This DNS tthen asks the authoritative DNS server (in this case run by its CDN Akamai), which trys to choose the closest server. Since the CDN can only see the origin of the last request, it has to approximate your location based on the location of your DNS server.

When selecting replica servers, CDNs try to choose the server with the lowest latency, typically the closest geogaphically, to its intended destination. Latency is important factor for web performance since the typical content request include several round trip times before the user receives data, including a DNS resolution, a TCP connection and HTTP header exchange. The multiple round trips mean that any latency improvements to the content server are compounded throughout the transfer.

Conventional widsom in replica selection is that the closer you are to your DNS server, the closer your content servers will be. It is very hard to improve upon the DNS server of your internet service provider when it comes to replica server selection. For example, our previous investigation into the impact of Public DNS on replica server selection (paper here )found that replicas selected using public DNS resolves had over twice the latency when compared to those chosen by your ISP's DNS.

Replica selection in cellular networks seems to contradict these traditional beliefs. In Part 1 of this series, we saw that public DNS resolvers like GoogleDNS and OpenDNS are typcaily farther away from clients resulting in longer DNS resolution than the ISP configured resolver. However, when we look at the latency to the replica servers returned by each DNS, we see that public DNS resolvers returned servers which had equal or lower latency 74% of the time in the case of OpenDNS! Figure 1 below shows the percentage of time the replicas returned by public DNS servies had equal or improved latencies for all hostnames we tested. It is also interesting to note that both public DNS resolvers returned servers with a 20% latency improvement around 20% of the time, and a 40% improvement 10% of the time.

% of Replica Selections
(larger is better)
Percentage Improvement
Figure 1. Percentage of times each public DNS service returned equal or better replica servers to devices.

When we look at replica selection on a per domain level, we see that it varies depending on the hostname chosen, and on the CDN they employ. The greatest example of disparity we saw was with both and, where OpenDNS performed as well or better 95% of the time. Figure 2 shows the percentage of time replicas returned had latencies which had equal or greater performance than the Local DNS options. The figure displays the wide disparity between public DNS services as well as across different hostnames. It seems clear that the current system of replica selection for cellular networks is far from optimal.

Fraction of Replica Selections (%)
(larger is better)
Figure 2. Percentage of times each public DNS service returned equal or better replica servers to devices for each hostname.

We have released a tool called Namehelp Mobile which measures the performance of your ISP provided DNS service against several public DNS options. In addition, it also measures the role that your DNS service has with Content Delivery Networks (CDNs) in mobile networks, and how your ISP provided DNS service can actually deliver worse performance in many cases.