LDplayer: DNS Experimentation at Scale. Zhu, L. & Heidemann, J. Technical Report 722, USC/Information Sciences Institute, Marina del Rey, CA, USA, November, 2017. ![LDplayer: DNS Experimentation at Scale [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex DNS has evolved over the last 20 years, improving in security and privacy and broadening the kinds of applications it supports. However, this evolution has been slowed by the large installed base with a wide range of implementations that are slow to change. Changes need to be carefully planned, and their impact is difficult to model due to DNS optimizations, caching, and distributed operation. We suggest that \emphexperimentation at scale is needed to evaluate changes and speed DNS evolution. This paper presents LDplayer, a configurable, general-purpose DNS testbed that enables DNS experiments to scale in several dimensions: many zones, multiple levels of DNS hierarchy, high query rates, and diverse query sources. LDplayer provides high fidelity experiments while meeting these requirements through its distributed DNS query replay system, methods to rebuild the relevant DNS hierarchy from traces, and efficient emulation of this hierarchy of limited hardware. We show that a single DNS server can correctly emulate multiple independent levels of the DNS hierarchy while providing correct responses as if they were independent. We validate that our system can replay a DNS root traffic with tiny error ($±\,8\,ms$ quartiles in query timing and $±\,0.1%$ difference in query rate). We show that our system can replay queries at 87k queries/s, more than twice of a normal DNS Root traffic rate, maxing out one CPU core used by our customized DNS traffic generator. LDplayer's trace replay has the unique ability to evaluate important design questions with confidence that we capture the interplay of caching, timeouts, and resource constraints. As an example, we can demonstrate the memory requirements of a DNS root server with all traffic running over TCP, and we identified performance discontinuities in latency as a function of client RTT.
@TechReport{Zhu17d,
author = "Liang Zhu and John Heidemann",
title = "LDplayer: {DNS} Experimentation at Scale",
institution = "USC/Information Sciences Institute",
year = 2017,
number = 722,
address = "Marina del Rey, CA, USA",
month = nov,
keywords = "Domain Name System (DNS), experiments, performance, trace replay",
sortdate = "2017-11-11",
project = "ant, retrofuturebridge, lacrend, nipet, researchroot",
jsubject = "dns",
jlocation = "johnh: pafile",
url = "https://ant.isi.edu/%7ejohnh/PAPERS/Zhu17d.html",
pdfurl = "https://ant.isi.edu/%7ejohnh/PAPERS/Zhu17d.pdf",
blogurl = "https://ant.isi.edu/blog/?p=1131",
codeurl = "https://ant.isi.edu/software/ldplayer/index.html",
myorganization = "USC/Information Sciences Institute",
copyrightholder = "authors",
abstract = "DNS has evolved over the last 20 years, improving in security and
privacy and broadening the kinds of applications it supports.
However, this evolution has been slowed by the large installed base
with a wide range of implementations that are slow to change. Changes
need to be carefully planned, and their impact is difficult to model
due to DNS optimizations, caching, and distributed operation. We
suggest that \emph{experimentation at scale} is needed to evaluate
changes and speed DNS evolution. This paper presents LDplayer, a
configurable, general-purpose DNS testbed that enables DNS experiments
to scale in several dimensions: many zones, multiple levels of DNS
hierarchy, high query rates, and diverse query sources. LDplayer
provides high fidelity experiments while meeting these requirements
through its distributed DNS query replay system, methods to rebuild
the relevant DNS hierarchy from traces, and efficient emulation of
this hierarchy of limited hardware. We show that a single DNS server
can correctly emulate multiple independent levels of the DNS hierarchy
while providing correct responses as if they were independent. We
validate that our system can replay a DNS root traffic with tiny error
($\pm\,8\,ms$ quartiles in query timing and $\pm\,0.1\%$ difference in
query rate). We show that our system can replay queries at 87k
queries/s, more than twice of a normal DNS Root traffic rate, maxing
out one CPU core used by our customized DNS traffic generator.
LDplayer's trace replay has the unique ability to evaluate important
design questions with confidence that we capture the interplay of
caching, timeouts, and resource constraints. As an example, we can
demonstrate the memory requirements of a DNS root server with all
traffic running over TCP, and we identified performance
discontinuities in latency as a function of client RTT.",
}
Downloads: 0
{"_id":"DLnqh3fcHTgjtMX9J","bibbaseid":"zhu-heidemann-ldplayerdnsexperimentationatscale-2017","author_short":["Zhu, L.","Heidemann, J."],"bibdata":{"bibtype":"techreport","type":"techreport","author":[{"firstnames":["Liang"],"propositions":[],"lastnames":["Zhu"],"suffixes":[]},{"firstnames":["John"],"propositions":[],"lastnames":["Heidemann"],"suffixes":[]}],"title":"LDplayer: DNS Experimentation at Scale","institution":"USC/Information Sciences Institute","year":"2017","number":"722","address":"Marina del Rey, CA, USA","month":"November","keywords":"Domain Name System (DNS), experiments, performance, trace replay","sortdate":"2017-11-11","project":"ant, retrofuturebridge, lacrend, nipet, researchroot","jsubject":"dns","jlocation":"johnh: pafile","url":"https://ant.isi.edu/%7ejohnh/PAPERS/Zhu17d.html","pdfurl":"https://ant.isi.edu/%7ejohnh/PAPERS/Zhu17d.pdf","blogurl":"https://ant.isi.edu/blog/?p=1131","codeurl":"https://ant.isi.edu/software/ldplayer/index.html","myorganization":"USC/Information Sciences Institute","copyrightholder":"authors","abstract":"DNS has evolved over the last 20 years, improving in security and privacy and broadening the kinds of applications it supports. However, this evolution has been slowed by the large installed base with a wide range of implementations that are slow to change. Changes need to be carefully planned, and their impact is difficult to model due to DNS optimizations, caching, and distributed operation. We suggest that \\emphexperimentation at scale is needed to evaluate changes and speed DNS evolution. This paper presents LDplayer, a configurable, general-purpose DNS testbed that enables DNS experiments to scale in several dimensions: many zones, multiple levels of DNS hierarchy, high query rates, and diverse query sources. LDplayer provides high fidelity experiments while meeting these requirements through its distributed DNS query replay system, methods to rebuild the relevant DNS hierarchy from traces, and efficient emulation of this hierarchy of limited hardware. We show that a single DNS server can correctly emulate multiple independent levels of the DNS hierarchy while providing correct responses as if they were independent. We validate that our system can replay a DNS root traffic with tiny error ($±\\,8\\,ms$ quartiles in query timing and $±\\,0.1%$ difference in query rate). We show that our system can replay queries at 87k queries/s, more than twice of a normal DNS Root traffic rate, maxing out one CPU core used by our customized DNS traffic generator. LDplayer's trace replay has the unique ability to evaluate important design questions with confidence that we capture the interplay of caching, timeouts, and resource constraints. As an example, we can demonstrate the memory requirements of a DNS root server with all traffic running over TCP, and we identified performance discontinuities in latency as a function of client RTT.","bibtex":"@TechReport{Zhu17d,\n\tauthor = \"Liang Zhu and John Heidemann\",\n\ttitle = \"LDplayer: {DNS} Experimentation at Scale\",\n institution = \"USC/Information Sciences Institute\",\n year = 2017,\n number = 722,\n address = \"Marina del Rey, CA, USA\",\n month = nov,\n\tkeywords = \"Domain Name System (DNS), experiments, performance, trace replay\",\n \tsortdate = \t\"2017-11-11\",\n\tproject = \"ant, retrofuturebridge, lacrend, nipet, researchroot\",\n\tjsubject = \"dns\",\n\tjlocation = \t\"johnh: pafile\",\n\turl =\t\t\"https://ant.isi.edu/%7ejohnh/PAPERS/Zhu17d.html\",\n\tpdfurl =\t\"https://ant.isi.edu/%7ejohnh/PAPERS/Zhu17d.pdf\",\n\tblogurl = \"https://ant.isi.edu/blog/?p=1131\",\n\tcodeurl = \t\"https://ant.isi.edu/software/ldplayer/index.html\",\n\tmyorganization =\t\"USC/Information Sciences Institute\",\n\tcopyrightholder = \"authors\",\n\tabstract = \"DNS has evolved over the last 20 years, improving in security and\nprivacy and broadening the kinds of applications it supports.\nHowever, this evolution has been slowed by the large installed base\nwith a wide range of implementations that are slow to change. Changes\nneed to be carefully planned, and their impact is difficult to model\ndue to DNS optimizations, caching, and distributed operation. We\nsuggest that \\emph{experimentation at scale} is needed to evaluate\nchanges and speed DNS evolution. This paper presents LDplayer, a\nconfigurable, general-purpose DNS testbed that enables DNS experiments\nto scale in several dimensions: many zones, multiple levels of DNS\nhierarchy, high query rates, and diverse query sources. LDplayer\nprovides high fidelity experiments while meeting these requirements\nthrough its distributed DNS query replay system, methods to rebuild\nthe relevant DNS hierarchy from traces, and efficient emulation of\nthis hierarchy of limited hardware. We show that a single DNS server\ncan correctly emulate multiple independent levels of the DNS hierarchy\nwhile providing correct responses as if they were independent. We\nvalidate that our system can replay a DNS root traffic with tiny error\n($\\pm\\,8\\,ms$ quartiles in query timing and $\\pm\\,0.1\\%$ difference in\nquery rate). We show that our system can replay queries at 87k\nqueries/s, more than twice of a normal DNS Root traffic rate, maxing\nout one CPU core used by our customized DNS traffic generator. \nLDplayer's trace replay has the unique ability to evaluate important\ndesign questions with confidence that we capture the interplay of\ncaching, timeouts, and resource constraints. As an example, we can\ndemonstrate the memory requirements of a DNS root server with all\ntraffic running over TCP, and we identified performance\ndiscontinuities in latency as a function of client RTT.\",\n}\n\n","author_short":["Zhu, L.","Heidemann, J."],"bibbaseid":"zhu-heidemann-ldplayerdnsexperimentationatscale-2017","role":"author","urls":{"Paper":"https://ant.isi.edu/%7ejohnh/PAPERS/Zhu17d.html"},"keyword":["Domain Name System (DNS)","experiments","performance","trace replay"],"metadata":{"authorlinks":{}}},"bibtype":"techreport","biburl":"https://bibbase.org/f/dHevizJoWEhWowz8q/johnh-2023-2.bib","dataSources":["YLyu3mj3xsBeoqiHK","fLZcDgNSoSuatv6aX","fxEParwu2ZfurScPY","7nuQvtHTqKrLmgu99"],"keywords":["domain name system (dns)","experiments","performance","trace replay"],"search_terms":["ldplayer","dns","experimentation","scale","zhu","heidemann"],"title":"LDplayer: DNS Experimentation at Scale","year":2017}