Today, CITP updated the Interconnection Measurement Project website with the monthly data for January 2017 through March 2017. The site now includes eighteen months of data, starting in October 2015. Throughout the year and a half of data, aggregate peak utilization has remained below 50 percent, except in one instance where the monthly aggregate peak utilization reached just 51 percent (January 2016). Over the eighteen-month period, aggregate interconnection capacity has grown by 74 percent, while aggregate peak usage has grown by only 58 percent.
CITP continues to work closely with CableLabs, Deepfield, and the participating ISPs to monitor and ensure the quality and integrity of the data feed. As detailed in the blog post associated with the 4Q2016 update, Deepfield’s December updates to the measurement platforms of the participating ISPs caused significant outages to the data feed received by CITP. These outages continued in a meaningful way through mid-January.
With the continued data feed outages, we were unable to generate the full suite of analysis and measures for the month of January. In particular, we were unable to produce the detailed measures for certain regions because the outages caused the number of providers in certain regions to fall below three, preventing us from releasing measures for those regions. In addition, we were unable to produce a full heat map for the month of January due to the data feed outages. By mid-January, Deepfield was able to restore the data feed, including regional information, for the vast majority of interconnection capacity under measurement.
Unrelated to the data feed outages, we have also had to omit measures for the Phoenix region due to changes in interconnection arrangements. For the months of February and March, we no longer have three ISPs with interconnects in this region. It is not uncommon for an ISP to continue to refine its interconnection arrangements which may cause an ISP to pull out of a given region. This is particularly true for smaller regions and for ISPs that have a smaller presence in a given region. For the time being, we’ll continue to maintain the Phoenix regional breakout in the analysis, given that a third participating ISP may begin interconnecting in that region in the future, enabling renewed reporting of measures for Phoenix.
Given the continued evolution of the Internet, interconnection, and the underlying infrastructure, it isn’t surprising to see data interruptions or shifts in regional deployments. To ensure the quality and integrity of the measurement data and the overall project, we remain vigilant in working with our industry partners to continuously monitor the data we receive.