ASNs Missing in Action

Similar documents
MENOG 6 Policy Update and Resource Statistics

Routing Information Service Status and Plans

STATUS UPDATE. ARIN XII Chicago. 22 October Adiel AKPLOGAN

Unit 6: Probability Plotting

Clustering radar tracks to evaluate efficiency indicators Roland Winkler Annette Temme, Christoph Bösel, Rudolf Kruse

Hawker 800XP Versus Falcon 2000 SAMPLE

JETNET & ACC Reporting Sample Report - F900EX versus G500

Methodology and coverage of the survey. Background

Evaluation of Strategic and Tactical Runway Balancing*

Improving Taxi Boarding Efficiency at Changi Airport

Supporting the design of a state aviation safety plan

SATELLITE CAPACITY DIMENSIONING FOR IN-FLIGHT INTERNET SERVICES IN THE NORTH ATLANTIC REGION

Part 121 CERTIFICATION AND OPERATIONS: AIR

Annual & Hourly Cost Detail

Ticketing and Booking Data

CONNECT Events: Flight Optimization

LACNIC REPORT. RIPE 46 Amsterdam Netherlands. Raúl Echeberría CEO RIPE 46

ATTEND Analytical Tools To Evaluate Negotiation Difficulty

Price-Setting Auctions for Airport Slot Allocation: a Multi-Airport Case Study

Congestion Management Alternatives: a Toolbox Approach

A Primer on Fatigue Damage Spectrum for Accelerated and Reliability Testing

DGCA - CAR 7 - J - I FLIGHT DUTY TIME AND FLIGHT TIME LIMITATIONS CABIN CREW MEMBER. Quick Reference Guide Version 2.1

ANALYSIS OF THE CONTRIUBTION OF FLIGHTPLAN ROUTE SELECTION ON ENROUTE DELAYS USING RAMS

FIXED-SITE AMUSEMENT RIDE INJURY SURVEY, 2013 UPDATE. Prepared for International Association of Amusement Parks and Attractions Alexandria, VA

Evaluation of Predictability as a Performance Measure

LACNIC REPORT. Ricardo Patara RSG Manager lacnic.net. RIPE 49 September 2004 Manchester

Kristina Ricks ISYS 520 VBA Project Write-up Around the World

Falcon 2000S & Challenger 350 Analysis

Monthly Australian road deaths last five years, with trend. 60 Jan 08 Jan 09 Jan 10 Jan 11 Jan 12 Jan 13

CENTRAL OREGON REGIONAL TRANSIT MASTER PLAN

Thanksgiving Holiday Period Traffic Fatality Estimate, 2017

Atennea Air. The most comprehensive ERP software for operating & financial management of your airline

MARRIOTT INTERNATIONAL, INC. PRESS RELEASE SCHEDULES QUARTER 4, 2016 TABLE OF CONTENTS

Fuel Burn Impacts of Taxi-out Delay and their Implications for Gate-hold Benefits

Benefits Analysis of a Runway Balancing Decision-Support Tool

Defining housing markets using postcode sectors

Evaluation of Alternative Aircraft Types Dr. Peter Belobaba

GUIDE TO THE DETERMINATION OF HISTORIC PRECEDENCE FOR INNSBRUCK AIRPORT ON DAYS 6/7 IN A WINTER SEASON. Valid as of Winter period 2016/17

Analysis of en-route vertical flight efficiency

Treasure Island Supplemental Information Report Addendum

GATWICK NIGHT MOVEMENT AND QUOTA ALLOCATION PROCEDURES

Subject: Flight Duty Time and Flight Time Limitations Cabin Crew

A Model to Assess the Mobility of the National Airspace System (NAS).

Furthermore, both our Flight API and our WebFare Engine have improved largely and have been supplemented with new features.

1. Introduction. 2.2 Surface Movement Radar Data. 2.3 Determining Spot from Radar Data. 2. Data Sources and Processing. 2.1 SMAP and ODAP Data

Have Descents Really Become More Efficient? Presented by: Dan Howell and Rob Dean Date: 6/29/2017

Corporate Productivity Case Study

Table of Content. Table of Contents Mobile Experts LLC. All Rights Reserved. 1

Citation XLS Analysis - Owner & Charter Hour Contributions. Prepared March 31, 2017

ADMINISTRATIVE DATA SOURCE FOR ACCOMMODATION STATISTICS IN CROATIA

In-Service Data Program Helps Boeing Design, Build, and Support Airplanes

Learning Objectives 7.3 Flight Performance and Planning Flight Planning & Flight Monitoring

Measuring the Business of the NAS

The performance of Scotland s high growth companies

Facebook and the GFW of China. Geoff Huston, Byron Ellacott APNIC

CONSTELLATIONS' PERFORMANCE ANALYSIS AND SIMULATION FOR COCKPIT AND PASSENGERS AERONAUTICAL SERVICES

2nd Quarter. AEDC is pleased to present the Anchorage Quarterly Economic Indicators Report for the second quarter of 2010.

SAMTRANS TITLE VI STANDARDS AND POLICIES

Business Intelligence Development at Winnipeg Transit

IPv6 Trends in the RIPE NCC Service Region. Nathalie Trenaman 12 September 2016 IPv6 Day Israel

LACNIC REPORT. ARIN XI - Memphis, USA. Raúl Echeberría LACNIC CEO < > ARIN XI

Implementation of Real-Time Data into a Transport Simulation

HEATHROW COMMUNITY NOISE FORUM

Residential Property Price Index

Table of Contents 2015 Mobile Experts LLC. All Rights Reserved. 1

DMAN-SMAN-AMAN Optimisation at Milano Linate Airport

Citation Sovereign Analysis. 250 Owner Hours w/ 250 Charter Hour Contribution

Pre-Coordination Runway Scheduling Limits Winter 2014

ICTAP Program. Interoperable Communications Technical Assistance Program. Communication Assets Survey and Mapping (CASM) Tool Short Introduction

Annual & Hourly Cost Detail

Annual & Hourly Cost Detail

THINGS YOU FORGOT OR DIDN'T KNOW ABOUT EMPOWER

Airplane Value Analysis Alex Philip

Measure 67: Intermodality for people First page:

Why we need to compare wildlife strike data among airports to improve aviation safety

Annual & Hourly Cost Detail

Global Runway Safety Simposium

Guidance on criteria for assessing the financial resources of new applicants and holders of operating licences

Measuring Ground Delay Program Effectiveness Using the Rate Control Index. March 29, 2000

DGCA - CAR 7 - J - I FLIGHT DUTY TIME AND FLIGHT TIME LIMITATIONS CABIN CREW MEMBER. Quick Reference Guide Version 1.1

15:00 minutes of the scheduled arrival time. As a leader in aviation and air travel data insights, we are uniquely positioned to provide an

BusStop Telco 2.0 application supporting public transport in agglomerations

Introduction to Business Statistics I Homework # 2

Fair Allocation Concepts in Air Traffic Management

Simulation of disturbances and modelling of expected train passenger delays

The EUROCONTROL CNS dashboard - User Manual -

Depeaking Optimization of Air Traffic Systems

Washington Dulles International Airport (IAD) Aircraft Noise Contour Map Update. Ultimate Operations 5th Working Group Briefing 9/25/18

DGCA FDTL 2016: FLIGHT DUTY TIME AND FLIGHT TIME LIMITATIONS - CABIN CREW MEMBER

Latin America Regional Connectivity Update

GATWICK NIGHT MOVEMENT AND QUOTA ALLOCATION PROCEDURES

Management Presentation. November 2018

Perth Noise Abatement Procedures - Change to Preferred Runways

Adaptive Noise Cancellation using the LMS Algorithm. December 7, 1999

Management Presentation. May 2013

FLIGHT TRANSPORTATION LABORATORY REPORT R87-5 AN AIR TRAFFIC CONTROL SIMULATOR FOR THE EVALUATION OF FLOW MANAGEMENT STRATEGIES JAMES FRANKLIN BUTLER

Air Law, Regulation and Compliance Management

AeroACCESSTM IT IS YOUR JET! ANY TIME, ANY WHERE, ANY WAY! TM

UNITED STATES OF AMERICA DEPARTMENT OF TRANSPORTATION OFFICE OF THE SECRETARY WASHINGTON, D.C.

NWA Council 20. Second-Quarter 2008 Meeting

Transcription:

ASNs Missing in Action A detailed look into RIR statistics and RIS observations René Wilhelm New Projects Group RIPE NCC <wilhelm@ripe.net> 1

Motivation Routing table analyses: not all assigned ASN seen RIS report @ripe44: routing table growth in 2002 ~2/3 of new assignements ~1/3 missing in action? What is happening? Compare RIR statistics and RIS routing data 2

Data Selection - RIS (1) RIS = Routing Information Service 13 remote route collectors at exchanges, 450 peers most recent data (3 months) in a database all raw data availabe on web, starting july 1999. Extract ASN info from raw data for each day, process both updates and RIB dumps split ASpath into components flag every ASN seen, even if only appears in 1 update 3

Data Selection - RIS (2) Results are stored in relational database For each ASN seen (in updates or RIB dump): first day seen in RIS last day seen in RIS total number of days seen Total ASN count for each day in separate table 4.5 years covered: 18 August 2000 to 31 March 2005 4

ASNs in routing system: RIS vs CIDR-report RIS ASN cout consistent with weekly CIDR report 5

Data Selection RIRs (1) RIRs publish IP & ASN statistics via ftp for each ASN or ASN block: assigned date and country since june 2002 ASN statistics from all RIRs Can create historical ASN count from latest stats start with latest total, count back the days, subtract all ASNs assigned each day per day ASN count But stats files provide a snapshot of current state no info on returned IP/ASN resources Better to look at all old stats files start countback from 2002-06-1, earliest day of stats 6

ASNs in stats vs. countback from 2004-12-31 2002-06-01: countback falls 1200 ASNs short 7

Data Selection RIRs (2) Process all RIR stats files up to 2004-12-31 RIS data processed until 2005-03-31 3 months for very last assigned ASN to make it to RIS For each ASN, add following records in database first_stats, last_stats date of the first / last stats file listing ASN first_assign, last_assign assignment date in first / last stats file first_rir, last_rir RIR which assigned the ASN in first / last stats file 8

RIRs ASN count 2002-2005 LACNIC ASNs reappear ARIN bad debts LACNIC startup ERX to RIPE NCC (fixed in 2005 stats) ARIN RIPE NCC APNIC LACNIC APNIC corrections 9

Monthly ASN assignments & stats changes 10

RIRs, RIS & CIDR report ASN count 11

Fraction of assigned ASNs seen 12

Missing in Action 31 December 2004: Missing 11500 ASNs 30000 assigned ASNs in the RIR statistics files 18500 of those are seen in RIS Have all never been used on the Internet? No: 5000 were in RIS before but no longer seen (Q1 2005) 500 appeared for the first time in RIS in 2005 6000 assigned ASNs never seen in RIS 5 to 10% of 2004 assignments still expected but majority will not ever show up in global routing tables 13

Monthly RIR assignments, RIS observations 14

Time left for 16bit ASNs? Last three years more or less linear growth 270 +/- 25 new ASNs in RIS per month 105 +/- 15 ASN disappearing from RIS each month net growth 165 +/- 30 per month 300 +/- 40 RIR assignments per month if growth remains at these levels unseen ASN not reclaimed; returned ASN not reused: 33000 left RIRs run out in 8-10 years all unseen ASN reclaimed, all returned ASN reused: 46000 left RIRs run out in 20-28 years 15

Other statistics 16

Reassigned AS Numbers??? 17

Reassigned AS Numbers?? (2) Not really clear how to interpret the date changes Adminstrative reasons? e.g. transfer of ownership in ISP merger, take over Difficult to asses true age of these ASNs Remainder of the analysis: Only use ASNs with dates max 1 month apart ASN at least 7 days in RIS to be counted as seen (filter noise from unauthorized, accidental use of ASN) 18

Lifetime in RIS of ASNs no longer seen 4908 ASNs not seen for at least 3 months 35 ASNs from a block assigned to USAF - HSSC withdrawn at same time 19

Lifetime of ASNs assigned after 2000-08-18 20

Age distribution of ASNs in RIS on 2005-03-31 21

Activation delays How long does it take for an ASN to appear in the routing system after being assigned by a RIR? For each ASN assigned in 2001, 2002, 2003, 2004 calculate days between last assigned and first seen for each delay, count number of ASNs plot as cumulative fraction 22

Activation delay - 2001 assignments First seen in RIS 4 years (!) after assigned by RIR 23

Activation delay - 2002 assignments 24

Activation delay - 2003 assignments 25

Activation delay - 2004 assignments 26

Activation delay - ARIN assignments 27

Activation delay - RIPE NCC assignments 28

Activation delay - APNIC assignments 29

Activation delay - LACNIC assignments 30

CDF of ASNs leaving RIS - 2001 assignments 23% of ARIN s 2001 assignments no longer seen 31

CDF of ASNs leaving RIS - 2002 assignments 32

CDF of ASNs leaving RIS - 2003 assignments 33

CDF of ASNs leaving RIS - 2004 assignments 34

Conclusions Comparing RIR and RIS statistics provides interesting views on dynamics of ASN usage Two classes of Missing in Action ASNs once seen by RIS but stop being used ASNs assigned by RIR but never seen by RIS Can take year(s) for assigned ASN to appear in RIS If current growth rate continues, ASNs could get scarce as early as 2013. 35

36

37