Tag: SIGCOMM

Categories
Uncategorized

new conference paper “Anycast in Context: A Tale of Two Systems” at SIGCOMM 2021

We published a new paper “Anycast in Context: A Tale of Two Systems” by Thomas Koch, Ke Li, Calvin Ardi*, Ethan Katz-Bassett, Matt Calder**, and John Heidemann* (of Columbia, where not otherwise indicated, *USC/ISI, and **Microsoft and Columbia) at ACM SIGCOMM 2021.

From the abstract:

Anycast is used to serve content including web pages and DNS, and anycast deployments are growing. However, prior work examining root DNS suggests anycast deployments incur significant inflation, with users often routed to suboptimal sites. We reassess anycast performance, first extending prior analysis on inflation in the root DNS. We show that inflation is very common in root DNS, affecting more than 95% of users. However, we then show root DNS latency hardly matters to users because caching is so effective. These findings lead us to question: is inflation inherent to anycast, or can inflation be limited when it matters? To answer this question, we consider Microsoft’s anycast CDN serving latency-sensitive content. Here, latency matters orders of magnitude more than for root DNS. Perhaps because of this need, only 35% of CDN users experience any inflation, and the amount they experience is smaller than for root DNS. We show that CDN anycast latency has little inflation due to extensive peering and engineering. These results suggest prior claims of anycast inefficiency reflect experiments on a single application rather than anycast’s technical potential, and they demonstrate the importance of context when measuring system performance.

Tom also blogged about this work at APNIC.

Categories
News

the ns family of network simulators are awarded the SIGCOMM Networking Systems Award for 2020

From the SIGCOMM mailing list and Facebook feed, Dina Papagiannaki posted on June 3, 2020:

I hope that everyone in the community is safe. A brief announcement that we have our winner for the networking systems award 2020. The committee comprised of Anja Feldmann (Max-Planck-Institut für Informatik), Srinivasan Keshav (University of Cambridge, chair), and Nick McKeown (Stanford University) have decided to present the award to the ns family of network simulators (ns-1, ns-2, and ns-3). Congratulations to all the contributors!

Description

“ns” is a well-known acronym in networking research, referring to a series of network simulators (ns-1, ns-2, and ns-3) developed over the past twenty five years. ns-1 was developed at Lawrence Berkeley National Laboratory (LBNL) between 1995-97 based on an earlier simulator (REAL, written by S. Keshav). ns-2 was an early open source project, developed in the 1997-2004 timeframe and led by collaborators from USC Information Sciences Institute, LBNL, UC Berkeley, and Xerox PARC. A companion network animator (nam) was also developed during this time [Est00]. Between 2005-08, collaborators from the University of Washington, Inria Sophia Antipolis, Georgia Tech, and INESC TEC significantly rewrote the simulator to create ns-3, which continues today as an active open source project.

All of the ns simulators can be characterized as packet-level, discrete-event network simulators, with which users can build models of computer networks with varying levels of fidelity, in order to conduct performance evaluation studies. The core of all three versions is written in C++, and simulation scripts are written directly in a native programming language: for ns-1, in the Tool Command Language (Tcl), for ns-2, in object-oriented Tcl (OTcl), and for ns-3, in either C++ or Python. ns is a full-stack simulator, with a high degree of abstraction at the physical and application layers, and varying levels of modeling detail between the MAC and transport layers. ns-1 was released with a BSD software license, ns-2 with a collection of licenses later consolidated into a GNU GPLv2-compatible framework, and ns-3 with the GNU GPLv2 license.

ns-3 [Hen08, Ril10] can be viewed as a synthesis of three predecessor tools: yans [Lac06], GTNetS [Ril03], and ns-2 [Bre00]. ns-3 contains extensions to allow distributed execution on parallel processors, real-time scheduling with emulation capabilities for packet exchange with real systems, and a framework to allow C and C++ implementation (application and kernel) code to be compiled for reuse within ns-3 [Taz13]. Although ns-3 can be used as a general-purpose discrete-event simulator, and as a simulator for non-Internet-based networks, by far the most active use centers around Internet-based simulation studies, particularly those using its detailed models of Wi-Fi and 4G LTE systems. The project is now focused on developing models to allow ns-3 to support research and standardization activities involving several aspects of 5G NR, next-generation Wi-Fi, and the IETF Transport Area.

The ns-3-users Google Groups forum has over 9000 members (with several hundred monthly posts), and the developer mailing list contains over 1500 subscribers. Publication counts (as counted annually) in the ACM and IEEE digital libraries, as well as search results in Google Scholar, describing research work using or extending ns-2 and ns-3, continue to increase each year, and usage also appears to be growing within the networking industry and government laboratories. The project’s home page is at https://www.nsnam.org, and software development discussion is conducted on the ns-developers@isi.edu mailing list.

Nominees

The main authors of ns-1 were (in alphabetical order): Kevin Fall, Sally Floyd, Steve McCanne, and Kannan Varadhan.

ns-2 had a larger number of contributors. Space precludes listing all authors, but the following people were leading source code committers to ns-2 (in alphabetical order): Xuan Chen, Kevin Fall, Sally Floyd, Padma Haldar, John Heidemann, Tom Henderson, Polly Huang, K.C. Lan, Steve McCanne, Giao Ngyuen, Venkat Padmanabhan, Yuri Pryadkin, Kannan Varadhan, Ya Xu, and Haobo Yu. A more complete list of ns-2 contributors can be found at: https://www.isi.edu/nsnam/ns/CHANGES.html.

The ns-3 simulator has been developed by over 250 contributors over the past fifteen years. The original main development team consisted of (in alphabetical order): Raj Bhattacharjea, Gustavo Carneiro, Craig Dowell, Tom Henderson, Mathieu Lacage, and George Riley.

Recognition is also due to the long list of ns-3 software maintainers, many of which made significant contributions to ns-3, including (in alphabetical order): John Abraham, Zoraze Ali, Kirill Andreev, Abhijith Anilkumar, Stefano Avallone, Ghada Badawy Nicola Baldo, Peter D. Barnes, Jr., Biljana Bojovic, Pavel Boyko, Junling Bu, Elena Buchatskaya, Daniel Camara, Matthieu Coudron, Yufei Cheng, Ankit Deepak, Sebastien Deronne, Tom Goff, Federico Guerra, Budiarto Herman, Mohamed Amine Ismail, Sam Jansen, Konstantinos Katsaros, Joe Kopena, Alexander Krotov, Flavio Kubota, Daniel Lertpratchya, Faker Moatamri, Vedran Miletic, Marco Miozzo, Hemanth Narra, Natale Patriciello, Tommaso Pecorella, Josh Pelkey, Alina Quereilhac, Getachew Redieteab, Manuel Requena, Matias Richart, Lalith Suresh, Brian Swenson, Cristiano Tapparello, Adrian S.W. Tam, Hajime Tazaki, Frederic Urbani, Mitch Watrous, Florian Westphal, and Dizhi Zhou.

The full list of ns-3 authors is maintained in the AUTHORS file in the top-level source code directory, and full commit attributions can be found in the git commit logs.

References

[Bre00] Lee Breslau et al., Advances in network simulation, IEEE Computer, vol. 33, no. 5, pp. 59-67, May 2000.

[Est00] Deborah Estrin et al., Network Visualization with Nam, the VINT Network Animator, IEEE Computer, vol. 33, no.11, pp. 63-68, November 2000.

[Hen08] Thomas R. Henderson, Mathieu Lacage, and George F. Riley, Network simulations with the ns-3 simulator, In Proceedings of ACM Sigcomm Conference (demo), 2008.

[Lac06] Mathieu Lacage and Thomas R. Henderson. 2006. Yet another network simulator. In Proceeding from the 2006 workshop on ns-2: the IP network simulator (WNS2 ’06). Association for Computing Machinery, New York, NY, USA, 12–es.

[Ril03] George F. Riley, The Georgia Tech Network Simulator, In Proceedings of the ACM SIGCOMM Workshop on Models, Methods and Tools for Reproducible Network Research (MoMeTools) , Aug. 2003.

[Ril10] George F. Riley and Thomas Henderson, The ns-3 Network Simulator. In Modeling and Tools for Network Simulation, SpringerLink, 2010.

[Taz13] Hajime Tazaki et al. Direct code execution: revisiting library OS architecture for reproducible network experiments. In Proceedings of the ninth ACM conference on Emerging networking experiments and technologies (CoNEXT ’13). Association for Computing Machinery, New York, NY, USA, 217–228.

USC/ISI had multiple projects and was very active in ns-2 development for many years, first lead by Deborah Estrin (with the VINT project), then by John Heidemann (with the SAMAN and SCADDS projects), with Tom Henderson took over leadership and evolved it (with others) into ns-3. All of these efforts have been open source collaborations with key players at other institutions as well. Sally Floyd, Steve McCanne, and Kevin Fall were all leaders.

I would particularly like to thank the several USC students who did PhDs on ns-2 related topics: Polly Huang, Kun-Chan Lan, Debojyoti Dutta, and earlier Kannan Varadhan. Ns-2 also benefited from external code contributions from David B. Johnson’s Monarch group (then at CMU) and Elizabeth Belding and Charles Perkins. (My apologies for other contributors I’m sure I’m missing.)


A huge thanks to the ns-1 authors (Kannan Varadhan was a USC student at the time), and a huge thanks to the ns-3 authors for taking over maintainership and evolution and keeping it vibrant.

Categories
Publications Technical Report

new technical report “LDplayer: DNS Experimentation at Scale (abstract with poster)”

We released a new technical report “LDplayer: DNS Experimentation at Scale (abstract with poster)”, ISI-TR-721, available at https://www.isi.edu/publications/trpublic/pdfs/ISI-TR-721.pdf.

The poster abstract and poster (included as part of the technical report) appeared at the poster session at the SIGCOMM 2017 in August 2017 in Los Angeles, CA, USA.

From the abstract:

In the last 20 years the core of the Domain Name System (DNS) has improved in security and privacy, and DNS use broadened from name-to-address mapping to a critical roles in service discovery and anti-spam. However, protocol evolution and expansion of use has been slow because advances must consider a huge and diverse installed base. We suggest that experimentation at scale can fill this gap. To meet the need for experimentation at scale, this paper presents LDplayer, a configurable, general-purpose DNS testbed. LDplayer enables DNS experiments to scale in several dimensions: many zones, multiple levels of DNS hierarchy, high query rates, and diverse query sources. To meet these requirements while providing high fidelity experiments, LDplayer includes a distributed DNS query replay system and methods to rebuild the relevant DNS hierarchy from traces. 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 show the importance of our system to evaluate pressing DNS design questions, using it to evaluate changes in DNSSEC key size.

Categories
Papers Publications

new conference paper “Trinocular: Understanding Internet Reliability Through Adaptive Probing” in SIGCOMM 2013

The paper “Trinocular: Understanding Internet Reliability Through Adaptive Probing” was accepted by SIGCOMM’13 in Hong Kong, China (available at http://www.isi.edu/~johnh/PAPERS/Quan13c with cite and pdf, or direct pdf).

100% detection of outages one round or longer
100% detection of outages one round or longer (figure 3 from the paper)

From the abstract:

Natural and human factors cause Internet outages—from big events like Hurricane Sandy in 2012 and the Egyptian Internet shutdown in Jan. 2011 to small outages every day that go unpublicized. We describe Trinocular, an outage detection system that uses active probing to understand reliability of edge networks. Trinocular is principled: deriving a simple model of the Internet that captures the information pertinent to outages, and populating that model through long-term data, and learning current network state through ICMP probes. It is parsimonious, using Bayesian inference to determine how many probes are needed. On average, each Trinocular instance sends fewer than 20 probes per hour to each /24 network block under study, increasing Internet “background radiation” by less than 0.7%. Trinocular is also predictable and precise: we provide known precision in outage timing and duration. Probing in rounds of 11 minutes, we detect 100% of outages one round or longer, and estimate outage duration within one-half round. Since we require little traffic, a single machine can track 3.4M /24 IPv4 blocks, all of the Internet currently suitable for analysis. We show that our approach is significantly more accurate than the best current methods, with about one-third fewer false conclusions, and about 30% greater coverage at constant accuracy. We validate our approach using controlled experiments, use Trinocular to analyze two days of Internet outages observed from three sites, and re-analyze three years of existing data to develop trends for the Internet.

Citation: Lin Quan, John Heidemann and Yuri Pradkin. Trinocular: Understanding Internet Reliability Through Adaptive Probing. In Proceedings of the ACM SIGCOMM Conference. Hong Kong, China, ACM. August, 2013. <http://www.isi.edu/~johnh/PAPERS/Quan13c>.

Datasets (listed here) used in generating this paper are available or will be available before the conference presentation.

Categories
Papers Publications

new conference paper “Understanding Block-level Address Usage in the Visible Internet” at SIGCOMM

The paper “Understanding Block-level Address Usage in the Visible Internet” was accepted and presented at SIGCOMM’10 in New Delhi, India (available at http://www.isi.edu/~johnh/PAPERS/Cai10a.html).

From the abstract:

Although the Internet is widely used today, we have little information about the edge of the network. Decentralized management, firewalls, and sensitivity to probing prevent easy answers and make measurement difficult. Building on frequent ICMP probing of 1% of the Internet address space, we develop clustering and analysis methods to estimate how Internet addresses are used. We show that adjacent addresses often have similar characteristics and are used for similar purposes (61% of addresses we probe are consistent blocks of 64 neighbors or more). We then apply this block-level clustering to provide data to explore several open questions in how networks are managed. First, we provide information about how effectively network address blocks appear to be used, finding that a significant number of blocks are only lightly used (most addresses in about one-fifth of /24 blocks are in use less than 10% of the time), an important issue as the IPv4 address space nears full allocation. Second, we provide new measurements about dynamically managed address space, showing nearly 40% of /24 blocks appear to be dynamically allocated, and dynamic addressing is most widely used in countries more recent to the Internet (more than 80% in China, while less than 30% in the U.S.). Third, we distinguish blocks with low-bitrate last-hops and show that such blocks are often underutilized.

Citation: Xue Cai and John Heidemann. Understanding Block-level Address Usage in the Visible Internet. In Proceedings of the ACM SIGCOMM Conference , p. to appear. New Delhi, India, ACM. August, 2010. <http://www.isi.edu/~johnh/PAPERS/Cai10a.html>.