the unexpected responsiveness of internet hosts
play

The Unexpected Responsiveness of Internet Hosts Neil Spring Me - PowerPoint PPT Presentation

Dec 22, 2022 •415 likes •893 views

The Unexpected Responsiveness of Internet Hosts Neil Spring Me Measure the Internet to evaluate and justify protocols that increase network reliability. Thesis work - measuring how routers are connected in practice to evaluate and

  1. The Unexpected Responsiveness of Internet Hosts Neil Spring

  2. Me • Measure the Internet to evaluate and justify protocols that increase network reliability. • Thesis work - measuring how routers are connected in practice to evaluate and enhance routing protocols in terms of how they exploit common network designs in routing around failures. • Recent work - measuring when residential links fail to determine how people and protocols should respond to faults.

  3. Residential Link Reliability • Residential links are: • Important: VoIP/911, Security cameras, Thermostats • Vulnerable: Exposed to 
 weather, loss of power, 
 singly-connected photo credit: Ode Street Tribune

  4. It’s Personal

  5. What I mean by “how … to respond to faults” • Small-scale individual questions: • Should I get more than one provider? Or change? • Is it just me? • System builder questions: • Would it help to coordinate with neighbors for mutual backup? • What fraction of errors can “Network Diagnostics” diagnose? • Policy questions: • Do cities with more buried wiring fare better or worse? • How does Maryland compare to Virginia, North America to Europe?

  6. How to detect network failures • “ping” is the fundamental tool. • Innocuous packets that have only one purpose (excuse me, are you alive?) • A response shows that the recipient is reachable and alive.

  7. No response ⇏ failure • IP service allows four bad things to happen to your packets: delay, duplication, corruption, and loss. • A lost echo request (are you there?) or reply (I sure am!) should happen 1-3% of the time without major failure.

  8. ThunderPing • 1. Watch for severe weather alert forecasts • 2. Ping addresses thought to be in that region before, during, and after the alert ☀ .3% • 3. Figure out if there actually ☁ .4% was weather, correlate failures ⛅ .3% with conditions ⛈ 2.0% ⛄ 3.0%

  9. Lost pings ⇒ outages 120 10 vantage points 100 80 RTT (ms) 60 40 20 0 0 2 4 6 8 10 time (hour)

  11. Failures in weather Charter Ameritech Speakeasy WildBlue Comcast CenturyLink Windstream Verizon FiOS Cox MegaPath Verizon DSL 0.5 UP ➡ DOWN rate relative to total rate 0.4 0.3 0.2 0.1 0.0 Clear Cloudy Fog Rain T-storm

  12. Some lost pings ⇒ ?? UP UP ??? 120 10 vantage points 100 80 RTT (ms) 60 40 20 0 0 1 2 time (hour)

  13. Two Questions • Could high delay create false outages? • Could renumbering cause false outages and alter their duration?

  14. When should pings time out?

  15. When should pings time out? Measurement platform Measurement platform Timeout (seconds) Timeout (seconds) RIPE Atlas RIPE Atlas 1 1 Scamper Scamper 2 (configurable) 2 (configurable) Hubble / iPlane Hubble / iPlane 2 (one retry) 2 (one retry) SamKnows SamKnows 3 3 Scriptroute / Thunderping Scriptroute / Thunderping 3 (configurable) 3 (configurable) ISI survey ISI survey 3 (collects all) 3 (collects all)

  16. Let’s confirm ~3s! • Dataset: ISI survey data: 1% of routed /24’s, pinged every 11 minutes. • Precise timing below 3s timeout. • Imprecise timing above 3s timeout. Any received echo reply is logged with time and source. • Approach: Look at all response times, including those longer than the timeout.

  17. Survey-detected RTTs 1.0 Fraction of addresses 0.8 Percentile of pings median 0.6 80th 90th 0.4 95th 98th 0.2 99th 0 0 2 4 6 RTT (Latency) (seconds) About 10% of addresses routinely respond after one second. About 10% of addresses routinely respond after one second. About 10% of addresses routinely respond after one second. The distribution appears clipped by the 3s limit. The distribution appears clipped by the 3s limit. The distribution appears clipped by the 3s limit.

  18. Transform Survey Data Reply Probe Source Time RTT Destination [1320291701.0] P v119 1.99.16.242 1.99.16.242 2960.995 45 [1320291701.0] P v119 1.99.16.242 1.99.16.242 2960.995 45 [1320291701.0] P v119 1.99.16.242 1.99.16.242 2960.995 45 [1320291701.0] P v119 1.99.16.242 1.99.16.242 2960.995 45 [1320291701.0] P v119 1.99.16.242 1.99.16.242 2960.995 45 [1320292364.0] P v119 1.99.16.242 1.99.16.242 2767.092 45 [1320292364.0] P v119 1.99.16.242 1.99.16.242 2767.092 45 [1320292364.0] P v119 1.99.16.242 1.99.16.242 2767.092 45 [1320292364.0] P v119 1.99.16.242 1.99.16.242 2767.092 45 [1320292364.0] P v119 1.99.16.242 1.99.16.242 2767.092 45 [1320293027.0] P v119 1.99.16.242 error_time_out [1320293027.0] P v119 1.99.16.242 error_time_out [1320293027.0] P v119 1.99.16.242 error_time_out [1320293027.0] P v119 1.99.16.242 error_time_out [1320293027.0] P v119 1.99.16.242 error_time_out [1320293031.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320293031.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320293031.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320293031.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320293031.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320293691.0] P v119 1.99.16.242 error_time_out [1320293691.0] P v119 1.99.16.242 error_time_out [1320293691.0] P v119 1.99.16.242 error_time_out [1320293691.0] P v119 1.99.16.242 error_time_out [1320293691.0] P v119 1.99.16.242 error_time_out [1320293696.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d005] [1320293696.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d005] [1320293696.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d005] [1320293696.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d005] [1320293696.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d005] [1320294354.0] P v119 1.99.16.242 error_time_out [1320294354.0] P v119 1.99.16.242 error_time_out [1320294354.0] P v119 1.99.16.242 error_time_out [1320294354.0] P v119 1.99.16.242 error_time_out [1320294354.0] P v119 1.99.16.242 error_time_out [1320294358.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320294358.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320294358.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320294358.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320294358.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d004] [1320295017.0] P v119 1.99.16.242 error_time_out [1320295017.0] P v119 1.99.16.242 error_time_out [1320295017.0] P v119 1.99.16.242 error_time_out [1320295017.0] P v119 1.99.16.242 error_time_out [1320295017.0] P v119 1.99.16.242 error_time_out [1320295030.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d013] [1320295030.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d013] [1320295030.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d013] [1320295030.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d013] [1320295030.0] P v119 no_probe_ip 1.99.16.242 0.000 45 [d013] [1320291701.0] P v119 1.99.16.242 1.99.16.242 2960.995 45 [1320291701.0] P v119 1.99.16.242 1.99.16.242 2960.995 45 [1320292364.0] P v119 1.99.16.242 1.99.16.242 2767.092 45 [1320292364.0] P v119 1.99.16.242 1.99.16.242 2767.092 45 [1320293027.0] P v119 1.99.16.242 1.99.16.242 4000.0000 45 [1320293027.0] P v119 1.99.16.242 1.99.16.242 4000.0000 45 [1320293691.0] P v119 1.99.16.242 1.99.16.242 5000.0000 45 [1320293691.0] P v119 1.99.16.242 1.99.16.242 5000.0000 45 [1320294354.0] P v119 1.99.16.242 1.99.16.242 4000.0000 45 [1320294354.0] P v119 1.99.16.242 1.99.16.242 4000.0000 45 [1320295017.0] P v119 1.99.16.242 1.99.16.242 13000.0000 45 [1320295017.0] P v119 1.99.16.242 1.99.16.242 13000.0000 45

  19. Absurdly long RTTs Percentile of pings Unexpected responses caused by broadcast and duplicate responses

  20. Filtering broadcast responses removes modes 1.0 Fraction of addresses Percentile of pings median 80th 90th 0.99 95th 98th 99th 1% of pings from 1% of addresses have RTTs > 145s 0.98 0 200 400 600 RTT (Latency) (seconds)

Recommend

Responsiveness Human perception Responsiveness in Java Responsiveness in Web Apps

Responsiveness Human perception Responsiveness in Java Responsiveness in Web Apps

Responsiveness Human perception Responsiveness in Java Responsiveness in Web Apps Chapter 14 1 CS349 -- Responsiveness Responsiveness 2 CS349 -- Responsiveness Responsive Applications User experience is affected by how

796 views • 53 slides
Measuring Internet filters and their unexpected impact Leonid Evdokimov nullcon, Goa, March

Measuring Internet filters and their unexpected impact Leonid Evdokimov nullcon, Goa, March

Measuring Internet filters and their unexpected impact Leonid Evdokimov nullcon, Goa, March 2018 WHAT IS I N T E R N E T C E N S O R S H I P? Internet Censorship |ntnt snsrp| 1. a distortion of the reality of the Internet

327 views • 31 slides
Responsive Web Design Niels Olof Bouvin 1 Overview Responsiveness? Responsiveness through

Responsive Web Design Niels Olof Bouvin 1 Overview Responsiveness? Responsiveness through

Responsive Web Design Niels Olof Bouvin 1 Overview Responsiveness? Responsiveness through scripting Events in the Web browser Web browser APIs The Number Guessing Game Talking to servers Responsiveness through styling Maps in the Browser

1.19k views • 77 slides
Fluxo Improving the Responsiveness of Internet Services with Automa7c Cache Placement Alexander

Fluxo Improving the Responsiveness of Internet Services with Automa7c Cache Placement Alexander

Fluxo Improving the Responsiveness of Internet Services with Automa7c Cache Placement Alexander Rasmussen UCSD (Presen7ng) Emre Kiciman MSR Redmond Benjamin Livshits MSR Redmond Madanlal Musuvathi MSR Redmond 1 Caching in

990 views • 62 slides
Internet Protocol (IP) Internet Protocol (IP) TCP versus UDP TCP versus UDP Low Low-

Internet Protocol (IP) Internet Protocol (IP) TCP versus UDP TCP versus UDP Low Low-

Special Course on Networked Virtual January 30, 2004 Environments Internet Protocol (IP) Internet Protocol (IP) TCP versus UDP TCP versus UDP Low Low- -level protocols used by hosts and routers level protocols used by hosts and routers

275 views • 5 slides
King : Estimating latency between arbitrary Internet end hosts Krishna Gummadi, Stefan Saroiu

King : Estimating latency between arbitrary Internet end hosts Krishna Gummadi, Stefan Saroiu

King : Estimating latency between arbitrary Internet end hosts Krishna Gummadi, Stefan Saroiu Steven D. Gribble University of Washington A Question: What can we do with a tool that could estimate latency between any two arbitrary hosts

621 views • 27 slides
SAME Presentation on Cybersecurity Andy Knauf CIO Mead & Hunt INTERNET SECURITY TESTING

SAME Presentation on Cybersecurity Andy Knauf CIO Mead & Hunt INTERNET SECURITY TESTING

SAME Presentation on Cybersecurity Andy Knauf CIO Mead & Hunt INTERNET SECURITY TESTING The engineers will scan Internet-visible hosts, identify services running on the hosts, and conduct testing for vulnerabilities to known exploits.

671 views • 15 slides
Assessing the geographic resolution of exhaustive tabulation for geolocating Internet hosts S.

Assessing the geographic resolution of exhaustive tabulation for geolocating Internet hosts S.

Assessing the geographic resolution of exhaustive tabulation for geolocating Internet hosts S. Siwpersad, S. Uhlig B. Gueye Delft University of Technology Universite de Liege The Netherlands Belgium 1 Vermelding onderdeel organisatie

408 views • 24 slides
SIGNS domain: Measuring responsiveness & Measuring responsiveness & MCID of TIS Mandy

SIGNS domain: Measuring responsiveness & Measuring responsiveness & MCID of TIS Mandy

SIGNS domain: Measuring responsiveness & Measuring responsiveness & MCID of TIS Mandy Schram, MD PhD Mandy Schram, MD PhD Aim To demonstrate the responsiveness and minimal clinically important difference of of Three Item

222 views • 19 slides
Introduction What is the Internet? Define network edge: hosts, access net, physical media

Introduction What is the Internet? Define network edge: hosts, access net, physical media

Computer Networks Chapter 1: Introduction CSC 249 January 25, 2018 Introduction What is the Internet? Define network edge: hosts, access net, physical media Define the network core & Internet structure First glimpse at the

414 views • 18 slides
Demand Responsiveness from the Rand HIE Demand Responsiveness from the Rand HIE Mannings et al,

Demand Responsiveness from the Rand HIE Demand Responsiveness from the Rand HIE Mannings et al,

Demand Responsiveness from the Rand HIE Demand Responsiveness from the Rand HIE Mannings et al, 1987 Newhouse, 1993 Dollars (1984$) Dollars (1984$) Total Cost Outpatient Inpatient Probability Health Cost Dental Preventi Office Mental

406 views • 3 slides
TCP/IP CIS 218/238 Internet Protocol (IP) The Internet Protocol (IP) is responsible for

TCP/IP CIS 218/238 Internet Protocol (IP) The Internet Protocol (IP) is responsible for

TCP/IP CIS 218/238 Internet Protocol (IP) The Internet Protocol (IP) is responsible for ensuring that data is transferred between two Intenret hosts based on a 32 bit address. To be ROUTABLE, a protocol must specify a NETWORK ADDRESS

899 views • 22 slides
Dialogue Managing the Unexpected Managing the Unexpected the Role of the Regulatory Body

Dialogue Managing the Unexpected Managing the Unexpected the Role of the Regulatory Body

Swiss Federal Nuclear Safety Inspectorate ENSI Dialogue Managing the Unexpected Managing the Unexpected the Role of the Regulatory Body Cornelia Ryser, ENSI, Switzerland IAEA Technical Meeting Vienna, 25 29 June 2012 Assumptions

543 views • 4 slides
Curriculum responsiveness and student employability in VET Bridge Post-School Community of

Curriculum responsiveness and student employability in VET Bridge Post-School Community of

Curriculum responsiveness and student employability in VET Bridge Post-School Community of Practice October 2016 Available at: www.lmip.org.za Outline Employability and Responsiveness: Opening up the debate Theme 4: Project 2 an

518 views • 24 slides
Programmers View of Internet Programmers View of Internet CS 105 Tour of the Black

Programmers View of Internet Programmers View of Internet CS 105 Tour of the Black

Programmers View of Internet Programmers View of Internet CS 105 Tour of the Black Holes of Computing 1. Hosts are mapped to a set of 32-bit IP(v4) addresses or 128-bit IP(v6)

314 views • 12 slides
Dynamic DNS: A Place To Call Home February 2, 2006 (With Help From Wikipedia) Hosts On The

Dynamic DNS: A Place To Call Home February 2, 2006 (With Help From Wikipedia) Hosts On The

NDLUG Dynamic DNS: A Place To Call Home February 2, 2006 (With Help From Wikipedia) Hosts On The Internet Every host has a unique IP Address 129.74.221.235 66.254.242.255 However, we use Domain Name System (DNS)

1.14k views • 9 slides
CS519: Computer Networks Lecture 4, Part 5: Mar 1, 2004 Internet Routing: ASs, igp, and BGP

CS519: Computer Networks Lecture 4, Part 5: Mar 1, 2004 Internet Routing: ASs, igp, and BGP

CS519: Computer Networks Lecture 4, Part 5: Mar 1, 2004 Internet Routing: ASs, igp, and BGP CS519 As we said earlier, the Internet is composed of Autonomous Systems (ASs) Where each AS is a set of routers, links, and hosts And

504 views • 36 slides
Online Meetings with Zoom For Participants and Hosts 1 Zoom for Participants and Hosts July

Online Meetings with Zoom For Participants and Hosts 1 Zoom for Participants and Hosts July

Zoom for Participants and Hosts July 2018 Online Meetings with Zoom For Participants and Hosts 1 Zoom for Participants and Hosts July 2018 What is Zoom? This Web Conferencing service is offered free of charge to eligible officers of

180 views • 17 slides
Pitfalls in PID: Expect the unexpected Faculty Disclosure X No, nothing to disclose Yes, please

Pitfalls in PID: Expect the unexpected Faculty Disclosure X No, nothing to disclose Yes, please

Pitfalls in PID: Expect the unexpected Faculty Disclosure X No, nothing to disclose Yes, please specify: PITFALLS IN PID: EXPECT THE UNEXPECTED First daughter of non-consanguineous parents, no relevant family history. 2 months old

258 views • 10 slides
Chapter 1: Introduction Our goal: Overview: get feel and whats the Internet? t

Chapter 1: Introduction Our goal: Overview: get feel and whats the Internet? t

Chapter 1: Introduction Our goal: Overview: get feel and whats the Internet? t terminology i l whats a protocol? more depth, detail network edge: hosts, access later in course nets, physical media approach:

651 views • 46 slides
No Yes Do you use the Internet to do homework? Do you use the Internet to follow your

No Yes Do you use the Internet to do homework? Do you use the Internet to follow your

No Yes Do you use the Internet to do homework? Do you use the Internet to follow your favourite team? Do you use the internet to watch music videos? Do you use the Internet to read the news? Do you use the Internet to play games? Shared

997 views • 51 slides
Fluid responsiveness Fluid responsiveness Monitoring in Surgical Monitoring in Surgical and

Fluid responsiveness Fluid responsiveness Monitoring in Surgical Monitoring in Surgical and

Fluid responsiveness Fluid responsiveness Monitoring in Surgical Monitoring in Surgical and Critically Ill Patients and Critically Ill Patients Impact clinique de la Goal-directed-therapy Goal-directed-therapy Impact clinique de la Patrice

930 views • 45 slides
PTO Meeting/PTO Hosts PTO Meeting/PTO Hosts Six Standards of Effective Six Standards of

PTO Meeting/PTO Hosts PTO Meeting/PTO Hosts Six Standards of Effective Six Standards of

10/4/2018 PTO Meeting/PTO Hosts PTO Meeting/PTO Hosts Six Standards of Effective Six Standards of Effective Pedagogy Pedagogy PTO Meeting: White Oak PTO PTO Meeting: White Oak PTO PTO Hosts Presentation by: Sarah Van Duzer and

551 views • 12 slides
Computers use IP addresses. Why DNS Session 1: Fundamentals Old solution: hosts.txt do we need

Computers use IP addresses. Why DNS Session 1: Fundamentals Old solution: hosts.txt do we need

Computers use IP addresses. Why DNS Session 1: Fundamentals Old solution: hosts.txt do we need names? Easier for people to remember A centrally-maintained file, distributed to all hosts Computers may be moved between networks, in on

689 views • 5 slides

More recommend