<?xml version="1.0" encoding="UTF-8"?><rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcq="http://purl.org/dc/terms/"><records count="1" morepages="false" start="1" end="1"><record rownumber="1"><dc:product_type>Conference Paper</dc:product_type><dc:title>Understanding Partial Reachability in the Internet Core</dc:title><dc:creator>Baltra, Guillermo; Saluja, Tarang; Pradkin, Yuri; Heidemann, John</dc:creator><dc:corporate_author>ACM</dc:corporate_author><dc:editor/><dc:description>Routing strives to connect all the Internet, but compete:  political
pressure threatens routing fragmentation; architectural changes such
as private clouds, carrier-grade NAT, and firewalls make connectivity
conditional; and commercial disputes create partial reachability for
days or years.  This paper suggests \emph{persistent, partial
reachability is fundamental to the Internet} and an underexplored
problem.  We
first \emph{derive a conceptual definition of the Internet core}
based on connectivity, not authority.  We
identify \emph{peninsulas}: persistent, partial connectivity;
and \emph{islands}: when computers are partitioned from the Internet
core.  Second, we develop algorithms to observe each across the
Internet, and apply them to two existing measurement systems:
Trinocular, where 6 locations observe 5M networks frequently, and RIPE
Atlas, where 13k locations scan the DNS roots frequently.
Cross-validation shows our findings are stable over
\emph{three years of data},
and consistent with as few as 3 geographically-distributed
observers.  We validate peninsulas and islands against CAIDA Ark,
showing good recall (0.94) and bounding precision between 0.42 and
0.82.  Finally, our work has broad practical impact:  we show
that \emph{peninsulas are more common than Internet outages}.
Factoring out peninsulas and islands as noise can \emph{improve
existing measurement systems}; their ``noise''
is $5\times$ to $9.7\times$ larger than the
operational events in RIPE's DNSmon.  We
show that most peninsula events are routing transients (45\%), but
most peninsula-time (90\%) is due to a few (7\%) long-lived events.
Our work helps inform Internet policy and governance, with our neutral
definition showing no single country or organization can unilaterally
control the Internet core.</dc:description><dc:publisher>ACM New Ideas in Networked Systems (NINeS)</dc:publisher><dc:date>2026-02-10</dc:date><dc:nsf_par_id>10661262</dc:nsf_par_id><dc:journal_name/><dc:journal_volume/><dc:journal_issue/><dc:page_range_or_elocation/><dc:issn/><dc:isbn/><dc:doi>https://doi.org/</dc:doi><dcq:identifierAwardId>2007106; 2028279; 2212480; 2530698</dcq:identifierAwardId><dc:subject>internet outages</dc:subject><dc:subject>partial outages</dc:subject><dc:subject>network partition</dc:subject><dc:subject>islands</dc:subject><dc:subject>peninsulas</dc:subject><dc:version_number/><dc:location/><dc:rights/><dc:institution/><dc:sponsoring_org>National Science Foundation</dc:sponsoring_org></record></records></rdf:RDF>