A couple of possibly interesting stats.
Seeing as CIDRD is around the corner and there's been the usual questions on how many CIDR routes, "more specific"s, etc. I did a couple of stats based on a full routing table (actually taken from an Ebone router) from last night at 21:00 UTC. The first table focuses on the current classful routes being passed around by AS and the amount of gain CIDR would give *IF* we could aggregate as the AS level. It shows the number of nets per AS so we see who is announcing what in terms of BGP as well as what the number of CIDR nets would be advertised if we could aggregate at this level (I know we cant but does give an indication as is the best you can base things on without local information - one day when the routing registry is fully integrated with an aggregate registry this should of course be possible - well that's the theory at least). One thing also to note in the this list is there are FOUR unknown ASes or at least I cant look them up. If you are the proud owner of one of these please register them a.s.a.p. The second shows the actual "*real live*" CIDR routes in the production Internet plus showing where "more specific" routes exist and where not. It also shows who is originating these aggregates and what the number of "more specific"s are also being announced (if any). In some case where the possible (all based on classful possible at the moment) "more specific"s equals the actual "more specific"s this means actually generating more routes than in the old classful world (but I know you all know this). The interesting point here are some SPs are really going for it i.e. no "more specific" routes whatsoever. Or, they're not real production nets of course. Hope this gives you an idea of what ASes are doing what. See you in Seattle. --Tony. Table 1 +++++++ ASnum NetsNow NetsCIDR % Gain Description AS701 1177 545 53.7% ALTERNET-AS AS174 1142 786 31.2% NYSERNET-AS AS372 908 693 23.7% NSN-AMES-AS AS1717 731 499 31.7% RENATER AS19 606 475 21.6% CSS-DOMAIN AS786 574 274 52.3% The JANET IP Service AS560 546 407 25.5% NEARNET-EXT-AS AS1225 499 314 37.1% CICNET3-AS AS200 498 364 26.9% BARRNET-AS AS2551 493 241 51.1% NETCOM-AS AS1275 489 258 47.2% DFN-IP AS209 455 346 24.0% WESTNET-EAST AS1239 435 240 44.8% SprintLink AS600 417 220 47.2% OARNET-AS AS1740 417 299 28.3% CERFNET AS279 409 305 25.4% SURANET-AS-2 AS1270 370 189 48.9% EUnet/DE AS86 351 311 11.4% SURANET-AS AS114 341 281 17.6% SESQUINET-AS AS97 284 199 29.9% JVNC-AS AS1849 279 168 39.8% PIPEX, Public IP EXchange AS685 253 172 32.0% NWNET-AS AS204 241 194 19.5% PSCNET-AS AS517 232 121 47.8% XLINK-UKA AS271 206 93 54.9% BCNET-AS AS81 186 95 48.9% CONCERT AS137 185 118 36.2% GARR AS93 183 140 23.5% MIDNET-AS AS210 179 138 22.9% WESTNET-WEST AS601 167 146 12.6% CANET1-AS AS568 159 142 10.7% SUMNET-AS AS1290 159 102 35.8% EUNet GB AS297 154 124 19.5% NSN-UMD-AS AS2150 152 72 52.6% CSUNET-SW AS1250 150 74 50.7% SINGAPORE-AS AS2907 142 90 36.6% SINET-AS AS237 136 104 23.5% NSFNETTEST14-AS AS1899 126 106 15.9% Fnet, EUnet-France AS1257 126 99 21.4% SWIPnet Swedish IP Network AS1103 113 88 22.1% SURFnet IP AS813 109 77 29.4% CANET-ASN AS1887 103 48 53.4% NASK Polish Research and Academic Comput AS1836 102 73 28.4% EUnet/CH Autonomous System AS267 101 85 15.8% CICNET2-AS AS555 89 52 41.6% MSCNET-AS AS2018 84 67 20.2% UNINET-ZA AS559 83 59 28.9% SWITCH, Swiss Academic and Research Netw AS719 79 52 34.2% LANLINK autonomous system AS1902 79 53 32.9% CESNET backbone A - EBONE AS378 77 50 35.1% ILAN AS766 76 55 27.6% RedIRIS Autonomous System AS22 76 57 25.0% NOSC AS1329 75 27 64.0% NSFNETTEST34-AS AS1901 73 52 28.8% EUnet Austria AS1955 72 34 52.8% Computer and Automation Institute AS2116 69 38 44.9% EUnet Norway AS1237 69 34 50.7% EUNSOL-AS AS2118 68 45 33.8% RIPE-ASNBLOCK2 AS2044 68 36 47.1% WORLDNET-AS AS109 68 43 36.8% CISCOSYSTEMS AS1267 63 48 23.8% I2U Autonomous System AS591 62 44 29.0% NSTN-AS AS376 62 36 41.9% RISQ-AS AS2548 62 41 33.9% DIGEX-AS AS293 61 53 13.1% ESNET-WEST-AS AS1890 60 38 36.7% NLnet Autonomous System AS2572 55 41 25.5% MORENET AS1204 55 34 38.2% SUNYNET-ASN-AS AS224 54 37 31.5% UNINETT, The Norwegian University & Rese AS2109 53 31 41.5% DKnet, EUnet Denmark AS1930 53 46 13.2% RCCN-NET AS2497 51 40 21.6% JNIC-ASN AS2110 51 23 54.9% IEunet (EUnet/Ireland) backbone AS AS1323 48 39 18.8% NSFNETTEST28-AS AS1251 48 41 14.6% ANBR-AS AS1741 46 38 17.4% FUNET autonomous system AS790 45 33 26.7% EUnet Finland AS2493 45 27 40.0% FONOROLA-EAST AS2119 45 30 33.3% TelePost IP net AS760 44 36 18.2% University of Vienna, Austria AS1241 44 24 45.5% FORTHnet Greek IP Service Provider AS542 39 30 23.1% ARNET AS1759 37 26 29.7% The DataNet IP Service AS1332 37 34 8.1% NSFNETTEST37-AS AS1982 35 29 17.1% ASN-NWNEXUS AS1729 34 33 2.9% TIPnet - Unidata IP services AS1331 34 23 32.4% NSFNETTEST36-AS AS2380 32 17 46.9% ESTNET Autonomous System AS789 31 25 19.4% IN2P3 AS2915 30 26 13.3% SPIN-NET AS2048 30 29 3.3% LANET-1 AS1850 30 28 6.7% ISnet Autonomous system AS2917 26 10 61.5% oleANe, Groupe APYSOFT AS2895 26 14 46.2% FREEnet AS2529 26 17 34.6% Demon Internet Ltd AS1328 26 24 7.7% NSFNETTEST33-AS AS1213 26 23 11.5% HEAnet autonomous system AS1967 25 16 36.0% TR-NET AS1324 25 23 8.0% NSFNETTEST29-AS AS803 24 23 4.2% SASK-NET-AS AS544 24 21 12.5% The DataNet IP Service AS2107 24 17 29.2% The ARNES IP Service AS2277 23 20 13.0% ECUANET AS1312 23 20 13.0% VA-TECH-AS AS2836 21 19 9.5% SUNET-STHLM AS2568 21 12 42.9% CTSNET AS2386 21 15 28.6% INS-AS AS771 20 20 0.0% NSN-RICE-AS AS684 20 17 15.0% MBNET-AS AS3130 20 9 55.0% RAIN-ASN AS2611 20 13 35.0% AS for BELNET sites connected to Brussel AS2111 20 18 10.0% AS for BELNET sites connected to Leuven AS1654 20 10 50.0% NORDUnet Iceland, Poland and Baltic AS2819 19 12 36.8% EUnet Czechia AS AS2702 19 12 36.8% INTERSERVE AS1705 19 12 36.8% GM-INDY-NET AS2828 18 10 44.4% INTERNEX AS278 18 15 16.7% RAM-AS AS2125 18 13 27.8% EUNET-WIN AS1327 18 18 0.0% NSFNETTEST32-AS AS286 17 12 29.4% EUnet-central AS AS2765 17 13 23.5% CONNECTED AS184 17 17 0.0% NSFNETTEST4-AS AS1752 17 13 23.5% BT main internal data network AS2902 16 12 25.0% WN-WY AS2637 16 13 18.8% GEORGIA-TECH AS2134 16 11 31.2% EUnet-SPAIN AS AS2120 16 7 56.2% DAXnet AS2043 16 13 18.8% European Multiprotocol Backbone AS1112 16 7 56.2% Montanuniversitaet Leoben, Austria AS35 15 13 13.3% MICATON AS2772 15 10 33.3% HKIGS AS1662 15 11 26.7% ANS-BLK1 AS1333 15 12 20.0% NSFNETTEST38-AS AS1205 15 7 53.3% Johannes Kepler University AS513 14 13 7.1% CERN/HEPNET AS2012 14 13 7.1% UNINET1 AS195 14 6 57.1% SDSC-AS AS1326 14 11 21.4% NSFNETTEST31-AS AS1274 14 10 28.6% GMD AS800 13 11 15.4% NLNET-AS AS2607 13 10 23.1% Slovak technical university AS2602 13 11 15.4% RESTENA AS2905 12 9 25.0% TICSA-ASN AS175 12 12 0.0% AFWL-AS AS68 11 11 0.0% LANL-INET-AS AS3380 11 6 45.5% PPPL-AS1 AS2704 11 9 18.2% HOOKUP-NET-A AS266 11 11 0.0% CICNET1-AS AS1227 11 11 0.0% SDSC-TEST1-AS AS683 10 10 0.0% ARGONNE-AS AS3463 10 10 0.0% ACES-AS AS2552 10 10 0.0% WUSTL-ASN AS1922 10 9 10.0% UAKOM-DAMIS AS714 9 7 22.2% APPLE-ENGINEERING-AS AS679 9 8 11.1% Technical University Vienna LAN AS3365 9 7 22.2% MCSNET AS2699 9 8 11.1% BBN-DSI1 AS2631 9 4 55.6% DISCOVERY-AS AS2136 9 7 22.2% PCC-TECHNO-AS AS1755 9 7 22.2% EBONE AS AS164 9 8 11.1% DDN-MB-AS AS1321 9 8 11.1% NSFNETTEST26-AS AS3403 8 8 0.0% TIAC AS233 8 7 12.5% NSFNETTEST10-AS AS1330 8 8 0.0% NSFNETTEST35-AS AS1294 8 7 12.5% PS-NETPLEX-AS AS698 7 6 14.3% ASN-UIUC-REGION-AS AS577 7 6 14.3% CA-NET-AS AS3404 7 7 0.0% COLORADOCOOP AS2847 7 6 14.3% LITNET, Lithuanian Academic and Research AS2108 7 7 0.0% Croatioan Academic and Research Network AS17 7 6 14.3% PURDUE-CS AS1322 7 7 0.0% NSFNETTEST27-AS AS3113 6 4 33.3% DOC-ESA-ASN AS2933 6 6 0.0% REALTIME AS2856 6 3 50.0% BT-UK-AS AS1853 6 4 33.3% ACOnet Backbone AS1842 6 6 0.0% USGS-AS AS1835 6 5 16.7% DENet - Danish Network for Research and AS1262 6 6 0.0% NSN-NCAR-AS-AS AS1104 6 6 0.0% NIKHEF-H AS73 5 5 0.0% WASHINGTON-AS AS3428 5 5 0.0% ESNET-AS AS3407 5 4 20.0% INTERPATH-NC AS3133 5 4 20.0% DHHS-NET-ASN AS292 5 5 0.0% ESNET-CENTRAL-AS AS2848 5 3 40.0% MSU.SU Autonomous system AS2715 5 5 0.0% REDERIO-AS AS2609 5 4 20.0% ASNBLK-RIPE AS26 5 5 0.0% CORNELL AS2276 5 5 0.0% OCS-AS1 AS194 5 4 20.0% USAN-AS AS1880 5 4 20.0% Stupi, house man's policy AS1803 5 2 60.0% ICMNET-5 AS1238 5 4 20.0% ICM-MALAYSIA AS1206 5 5 0.0% PSCNET-HS-AS AS1115 5 2 60.0% Austrian Academy of Sciences AS377 4 4 0.0% SNLA-NET-AS AS3111 4 4 0.0% DIRECT AS2927 4 3 25.0% ZONE-NET1 AS2900 4 4 0.0% WN-AZ AS2845 4 4 0.0% SUNET-LUND AS2841 4 4 0.0% SUNET-CTH AS2840 4 4 0.0% SUNET-GBG AS2832 4 4 0.0% SUNET-UMEA AS2820 4 4 0.0% Zelenograd Networks Community AS2605 4 4 0.0% CESNET backbone B - EMPB AS2578 4 4 0.0% RELCOM-DEMOS AS2555 4 4 0.0% INTER-GRP-B AS2146 4 4 0.0% CRNET AS2131 4 3 25.0% Goedel School Ges.m.b.H. AS1888 4 4 0.0% CWI-Amsterdam AS AS1784 4 4 0.0% SPRINTLINK AS1273 4 4 0.0% ECRC-RBS AS1234 4 3 25.0% IVOWAN-AS AS558 3 3 0.0% OLIVETTI-AS1 AS46 3 3 0.0% RUTGERS AS38 3 3 0.0% UIUC AS3381 3 3 0.0% MHPCC-AS AS3358 3 3 0.0% UCIA-AS AS3142 3 3 0.0% DORSAI AS3110 3 2 33.3% HOLONET AS3058 3 3 0.0% EMNET AS3 3 3 0.0% MIT-GATEWAYS AS2861 3 3 0.0% SNI-ZFE-AS AS2834 3 3 0.0% SUNET-UPPSALA AS2709 3 2 33.3% AOS-AS AS2706 3 3 0.0% HKSUPER AS2604 3 2 33.3% HS fuer Musik und darstellende Kunst, Gr AS1889 3 3 0.0% Hewlett-Packard Limited, AS1808 3 3 0.0% ICMNET-10 AS1800 3 3 0.0% ICM-Atlantic AS1776 3 2 33.3% ASN-WU-WIEN AS1747 3 3 0.0% IBMWATSON-AS AS1653 3 3 0.0% SUNET Swedish University Network AS1325 3 3 0.0% NSFNETTEST30-AS AS1292 3 3 0.0% ITESM-AS AS690 2 2 0.0% NSFNET-T3-RT-AS AS45 2 2 0.0% LLL-TIS-AS AS3447 2 2 0.0% CENTEL-CMDS AS3405 2 1 50.0% QNSNET AS3386 2 2 0.0% USNET-AS AS3134 2 2 0.0% USAID-ASN AS3127 2 2 0.0% CAPRICA-AS AS2843 2 2 0.0% SUNET-LINKOPING AS2831 2 2 0.0% SUNET-LULEA AS2643 2 2 0.0% IHEP-SU AS AS2599 2 1 50.0% SOVAM Teleport AS2561 2 2 0.0% Egyptian Networks AS2546 2 2 0.0% Ariadne Network (NCSR Demokritos) AS2047 2 2 0.0% KFKI-RMKI Autonomous system AS2038 2 2 0.0% IT-HEP Italian Hepnet AS2020 2 2 0.0% HWWILSON-AS AS2002 2 2 0.0% IBM-PV-AS AS1999 2 2 0.0% HOUSE-AS-1 AS1972 2 2 0.0% HPC-NET AS1782 2 2 0.0% UNKNOWN AS1754 2 2 0.0% DESY-HAMBURG AS1732 2 2 0.0% MIKROK-AS AS1660 2 2 0.0% ANS-BLK1 AS1133 2 1 50.0% DANTE/CERN T1 AS1125 2 2 0.0% SURFnet Test ASN AS9009 1 1 0.0% UNKNOWN AS9003 1 1 0.0% UNKNOWN AS9001 1 1 0.0% UNKNOWN AS8 1 1 0.0% RICE-AS AS603 1 1 0.0% CANET AS602 1 1 0.0% CANET2 AS565 1 1 0.0% VTTNET-AS AS549 1 1 0.0% ONET-AS AS545 1 1 0.0% NRI AS543 1 1 0.0% CONVEX AS3462 1 1 0.0% HINET-AS AS3461 1 1 0.0% GONZAGA-ASN AS3459 1 1 0.0% ONRAMP AS3425 1 1 0.0% ESNET-AS AS3397 1 1 0.0% EMI-AS AS3394 1 1 0.0% ELPRESS-AS AS3364 1 1 0.0% CSDCO-AS AS33 1 1 0.0% DEC-WRL-AS AS3152 1 1 0.0% FERMILAB AS3150 1 1 0.0% CLARKNET AS3139 1 1 0.0% ASEE AS3132 1 1 0.0% RCP-AS AS3114 1 1 0.0% DOJ-CRT-ASN AS2882 1 1 0.0% COREN-ONE AS288 1 1 0.0% ESA-AS AS2822 1 1 0.0% INNET BVBA AS2766 1 1 0.0% GLASNET AS2718 1 1 0.0% APSCN-AR AS2714 1 1 0.0% GSA_GOV AS2707 1 1 0.0% WEC AS2649 1 1 0.0% NAS2-AS AS2614 1 1 0.0% PUB AS2603 1 1 0.0% NORDUnet 2Mb AS2592 1 1 0.0% Turkish Universities and Research Center AS2588 1 1 0.0% Institute of Mathematics and CS AS2549 1 1 0.0% REDUDG AS2538 1 1 0.0% IBM-KINGS-AS AS2532 1 1 0.0% LOC AS2494 1 1 0.0% AKHNET Autonomous System AS2383 1 1 0.0% OTA AS238 1 1 0.0% NSFNETTEST15-AS AS2274 1 1 0.0% CBO AS2149 1 1 0.0% PSINET-2 AS2140 1 1 0.0% ISSC-AS AS2139 1 1 0.0% DOW1-AS AS2138 1 1 0.0% DOW2-AS AS2135 1 1 0.0% EuroKom Mail Service AS2128 1 1 0.0% DIT AS2112 1 1 0.0% RIPE-ASNBLOCK2 AS2049 1 1 0.0% AVL-GRAZ AS2036 1 1 0.0% JOANNEUM RESEARCH AS2028 1 1 0.0% BNR Europe AS2019 1 1 0.0% MORGAN-AS AS2004 1 1 0.0% TNO-HDO AS2003 1 1 0.0% EPRI-PA AS1997 1 1 0.0% IBMDES-AS AS1975 1 1 0.0% UMI-AS AS1957 1 1 0.0% ANSCIX-AS AS1956 1 1 0.0% IBMEVS-AS AS1934 1 1 0.0% ANS-RESTON-AS AS1882 1 1 0.0% Stupi, house woman's policy AS1804 1 1 0.0% ICMNET-6 AS1757 1 1 0.0% MDCI-AS AS1751 1 1 0.0% LEXMARK-AS AS1739 1 1 0.0% TAMNET AS1679 1 1 0.0% ANS-BLK1 AS1678 1 1 0.0% ANS-BLK1 AS1671 1 1 0.0% ANS-BLK1 AS1240 1 1 0.0% ICM-Pacific AS1224 1 1 0.0% NCSA-AS AS1134 1 1 0.0% PTT-Res Test AS1128 1 1 0.0% EuropaNET PoP Amsterdam AS1123 1 1 0.0% Technikum Vorarlberg AS1117 1 1 0.0% Universitaet fuer Bodenkultur, Wien AS1114 1 1 0.0% Graz University, Austria AS1113 1 1 0.0% Technische Universitaet Graz, AT AS1111 1 1 0.0% University of Klagenfurt, Austria AS1110 1 1 0.0% University of Innsbruck, Austria AS1109 1 1 0.0% AS Universitaet Salzburg Total 19400 13068 32.6% Table 2 +++++++ Explaination of Table 2 ------------------------ SR : Is there a "more specific" route advertised with the aggregate (Yes or No). P : Number of possible classful routes in aggregate. N : Actual "more specific" classful routes advertised. If P=N then actually advertising more routes rather than less. SR Aggregate Originator Start-Classful End-Classful P N -------------------------------------------------------------------------- Y 129.240.0.0/15 AS224 129.240 129.241 2 1 Y 130.232.0.0/15 AS1741 130.232 130.233 2 2 Y 192.36.2.0/23 AS1257 192.36.2 192.36.3 2 1 Y 192.36.18.0/23 AS1257 192.36.18 192.36.19 2 2 Y 192.36.220.0/23 AS1257 192.36.220 192.36.221 2 2 Y 192.71.152.0/23 AS1257 192.71.152 192.71.153 2 2 Y 192.87.108.0/22 AS1103 192.87.108 192.87.111 4 2 Y 192.87.136.0/22 AS1125 192.87.136 192.87.139 4 1 Y 192.91.242.0/23 AS513 192.91.242 192.91.243 2 2 N 192.108.196.0/23 AS1880 192.108.196 192.108.197 2 0 Y 192.108.200.0/21 AS1880 192.108.200 192.108.207 8 4 N 192.108.212.0/23 AS1880 192.108.212 192.108.213 2 0 N 192.121.64.0/23 AS1729 192.121.64 192.121.65 2 0 N 192.121.86.0/23 AS2836 192.121.86 192.121.87 2 0 Y 192.121.204.0/22 AS1257 192.121.204 192.121.207 4 4 Y 192.121.208.0/21 AS1257 192.121.208 192.121.215 8 8 Y 192.121.216.0/23 AS1257 192.121.216 192.121.217 2 2 Y 192.122.216.0/23 AS2110 192.122.216 192.122.217 2 2 Y 192.122.220.0/22 AS2110 192.122.220 192.122.223 4 4 N 192.150.56.0/21 AS1729 192.150.56 192.150.63 8 0 Y 192.150.64.0/20 AS1729 192.150.64 192.150.79 16 2 Y 192.150.80.0/21 AS1729 192.150.80 192.150.87 8 1 Y 192.216.0.0/16 AS200 192.216.0 192.216.255 256 115 Y 192.244.176.0/20 AS2497 192.244.176 192.244.191 16 8 Y 192.250.24.0/22 AS1890 192.250.24 192.250.27 4 4 Y 193.4.0.0/16 AS1850 193.4.0 193.4.255 256 29 N 193.10.6.0/23 AS2836 193.10.6 193.10.7 2 0 Y 193.12.54.0/23 AS1257 193.12.54 193.12.55 2 2 Y 193.12.56.0/23 AS1257 193.12.56 193.12.57 2 2 Y 193.12.94.0/23 AS1257 193.12.94 193.12.95 2 2 Y 193.12.106.0/23 AS1257 193.12.106 193.12.107 2 2 Y 193.12.118.0/23 AS1257 193.12.118 193.12.119 2 2 Y 193.12.122.0/23 AS1257 193.12.122 193.12.123 2 2 Y 193.12.128.0/22 AS1257 193.12.128 193.12.131 4 4 Y 193.12.150.0/23 AS1257 193.12.150 193.12.151 2 2 Y 193.44.0.0/16 AS1729 193.44.0 193.44.255 256 1 N 193.44.24.0/22 AS2836 193.44.24 193.44.27 4 0 Y 193.67.128.0/22 AS1890 193.67.128 193.67.131 4 4 Y 193.71.32.0/21 AS2116 193.71.32 193.71.39 8 7 Y 193.71.104.0/21 AS2116 193.71.104 193.71.111 8 8 Y 193.71.160.0/21 AS2116 193.71.160 193.71.167 8 8 Y 193.72.0.0/16 AS1836 193.72.0 193.72.255 256 56 Y 193.78.32.0/21 AS1890 193.78.32 193.78.39 8 8 Y 193.78.72.0/22 AS1890 193.78.72 193.78.75 4 3 Y 193.78.104.0/22 AS1890 193.78.104 193.78.107 4 4 Y 193.78.120.0/21 AS1890 193.78.120 193.78.127 8 6 Y 193.78.120.0/23 AS1890 193.78.120 193.78.121 2 2 Y 193.80.224.0/22 AS1901 193.80.224 193.80.227 4 2 Y 193.85.0.0/21 AS2819 193.85.0 193.85.7 8 4 Y 193.85.8.0/23 AS2819 193.85.8 193.85.9 2 2 Y 193.85.80.0/23 AS2819 193.85.80 193.85.81 2 2 Y 193.85.104.0/21 AS2819 193.85.104 193.85.111 8 3 Y 193.85.112.0/20 AS2819 193.85.112 193.85.127 16 4 Y 193.85.128.0/20 AS2819 193.85.128 193.85.143 16 2 Y 193.120.16.0/20 AS2110 193.120.16 193.120.31 16 16 Y 193.120.32.0/21 AS2110 193.120.32 193.120.39 8 8 Y 193.120.208.0/22 AS2110 193.120.208 193.120.211 4 2 Y 193.120.212.0/22 AS2110 193.120.212 193.120.215 4 2 Y 193.120.216.0/23 AS2110 193.120.216 193.120.217 2 2 Y 193.120.232.0/21 AS2110 193.120.232 193.120.239 8 1 Y 193.120.240.0/21 AS2110 193.120.240 193.120.247 8 2 N 193.128.40.0/21 AS1849 193.128.40 193.128.47 8 0 N 193.157.64.0/19 AS224 193.157.64 193.157.95 32 0 Y 193.166.0.0/16 AS1741 193.166.0 193.166.255 256 28 Y 193.170.54.0/23 AS1853 193.170.54 193.170.55 2 2 N 193.172.2.0/23 AS2043 193.172.2 193.172.3 2 0 Y 193.212.2.0/23 AS1257 193.212.2 193.212.3 2 2 Y 193.242.80.0/21 AS286 193.242.80 193.242.87 8 4 N 194.12.128.0/21 AS513 194.12.128 194.12.135 8 0 Y 198.30.0.0/16 AS600 198.30.0 198.30.255 256 165 Y 198.31.0.0/16 AS200 198.31.0 198.31.255 256 51 N 198.63.0.0/16 AS114 198.63.0 198.63.255 256 0 N 198.66.0.0/17 AS114 198.66.0 198.66.127 128 0 N 198.84.4.0/22 AS701 198.84.4 198.84.7 4 0 Y 198.92.0.0/17 AS109 198.92.0 198.92.127 128 46 N 198.93.160.0/19 AS109 198.93.160 198.93.191 32 0 Y 198.213.0.0/16 AS114 198.213.0 198.213.255 256 17 N 198.214.0.0/15 AS114 198.214.0 198.215.255 256 0 Y 198.216.0.0/16 AS114 198.216.0 198.216.255 256 4 Y 199.0.0.0/16 AS1239 199.0.0 199.0.255 256 64 N 199.0.192.0/20 AS1239 199.0.192 199.0.207 16 0 Y 199.1.0.0/16 AS1239 199.1.0 199.1.255 256 13 Y 199.2.0.0/16 AS1239 199.2.0 199.2.255 256 51 Y 199.3.0.0/16 AS1239 199.3.0 199.3.255 256 26 Y 199.18.0.0/16 AS600 199.18.0 199.18.255 256 65 N 199.108.0.0/16 AS1740 199.108.0 199.108.255 256 0 Y 202.24.208.0/20 AS2497 202.24.208 202.24.223 16 1 Y 202.32.0.0/16 AS2497 202.32.0 202.32.255 256 12
Oops - I've seen a mistake in Table 2. Line N 198.214.0.0/15 AS114 198.214.0 198.215.255 256 0 should read.. N 198.214.0.0/15 AS114 198.214.0 198.215.255 512 0 -Tony.
The first table focuses on the current classful routes being passed around by AS and the amount of gain CIDR would give *IF* we could aggregate as the AS level. It shows the number of nets per AS so we see who is announcing what in terms of BGP as well as what the number of CIDR nets would be advertised if we could aggregate at this level... Tony, Interesting. Just wonder if the would CIDR nets (aggregates) of each AS that you composed includes holes or not. E.g. If AS1 has routes 199.1.0 - 199.1.7 except 199.1.3. What CIDR net does it count for this AS in your list: a. 199.1.0/21 (aggregate with hole and reduce routes from 7 to 1) or b. 199.1.0, 199.1.1, 199.1.2 & 199.1.4/22 (without hole and reduce routes from 7 to 4) Basically, I like to know if this reduction of routes on your list (32.6%) is calculated with aggregates with holes or not. If it does not include holes, we should be able to see a MORE reduction of the routing table size by doing the same analysis but include holes in the potential aggregates. --Jessica
The first table focuses on the current classful routes being passed around by AS and the amount of gain CIDR would give *IF* we could aggregate as the AS level. It shows the number of nets per AS so we see who is announcing what in terms of BGP as well as what the number of CIDR nets would be advertised if we could aggregate at this level...
Tony,
Tony, On average each net learned by AS 690 is announced in 2.5 places. I think the range is typically 1-5 with the most common number being 2 (not sure on that). In any case, I think your assumption that we can aggregate at the AS level may be highly optimistic (at least for the near term - until we get far more sophisticated tools to track topology and aggregation implications). Of course, if CIDR either didn't prefer the more specific route but relied entirely on policy or punched holes in aggregates rather than overlap, this wouldn't be an issue, but that's the way CIDR works so it is an issue. It makes life much more interesting for Routing Authorities. :-) Curtis
On average each net learned by AS 690 is announced in 2.5 places.
This is an interesting peice of data, but does not tell us anything when we are okking at how much CIDR can by us. Looking at how many places the NSFNET sees inbound announcements of nets does not matter at all. All it tells you is that the global Internet is redundant in terms of links. Look at home ASs of nets - if you aggregate at the home AS, that is what is going to count. --asp@uunet.uu.net (Andrew Partan)
Looking at how many places the NSFNET sees inbound announcements of nets does not matter at all. All it tells you is that the global Internet is redundant in terms of links.
Look at home ASs of nets - if you aggregate at the home AS, that is what is going to count.
--asp@uunet.uu.net (Andrew Partan)
Andrew, A lot of the regional networks have more than one attachment point to ANSnet. Some use one as primary and one as backup. Most of the regional networks announce some of their networks as primary at one attachment point and other networks at another attachment point. This means that that regional could not aggregate before reaching ANSnet because to do so would mean losing the load balancing. One longer term way around this is to have ANS accept more specific routes plus an aggregate and propogate the more specific routes into it's IBGP but not propogate them further (just the aggregate). In any case, configuration is going to get a lot harder for many regionals to make sure that they aggregate equally to all their peers but still are capable of passing more specific routes to ANS or any peer network to which they attach at multiple places and for which they need to load balance. Knowing that the average is 2.5:1 just indicates that aggregating entire AS is not going to be easy and that the figures arrived at using that assumption are going to be very optimistic at least for the near term. The second point I made was that a major role of the routing arbitrator will be to provide enough knowlege of topology to make more aggregation possible so we can approach the kind of figures Tony projected. Maybe that point got lost because of the way it was presented. Regards, Curtis
If they have more than one attachement, then they are (currently) using more than one AS - so with "one AS one policy", they should be fine. [If folks are violating the "one AS one policy" rule, then they are going to have problems that they have brought upon themselves. BGP does not really support more than one AS per policy.] If there are 2 ASs that touch in more than one point, then BGP4's inter-AS metric should take care of this (you will need to have some more care in what routes you pass between these 2 ASs and what routes they pass on to their neighbors), but as you need to coordinate to have 2 ASs touch in more than one point anyhow, you should just need to add a more to the coordination. In any case, downstream ASs should not know or care or need to do anything to make this work. You certainly should be able to do aggregation by site w/o any problem (even for ASs that touch in more than 1 place).
The second point I made was that a major role of the routing arbitrator will be to provide enough knowlege of topology to make more aggregation possible so we can approach the kind of figures Tony projected. Maybe that point got lost because of the way it was presented.
Actaully, the role of the RA is to get people to give it their (local) data so that the RA can get a database that has (nearly) complete global info. Projections of how much we are going to get from CIDR will be somewhat soft until we started getting a couple of real life examples. I have poked at what we would get under various CIDR strategeies, but as I have not done it for "real", I am not 100% sure yet. Here is some results from a run that I did of my data a while back: 1381 total nets 646 CIDR routes if CIDRize by site 619 CIDR routes if CIDRize by AUP (NSFNET vs non-NSFNET) 1026 NSFNET nets; 468 NSFNET CIDR routes 355 non-NSFNET nets; 151 non-NSFNET CIDR routes 500 CIDR routes if CIDRize everything together Even if we just CIDRized by site, it looks like we will get nearly all of the savings that we would get if we fully CIDRized. --asp@uunet.uu.net (Andrew Partan)
If they have more than one attachement, then they are (currently) using more than one AS - so with "one AS one policy", they should be fine.
[If folks are violating the "one AS one policy" rule, then they are going to have problems that they have brought upon themselves. BGP does not really support more than one AS per policy.]
Where did this come from? I was talking about load balancing between predominantly T1 networks that touch the T3 network in two geographically distant places and require the T3 network to take an exit point that minimizes the slower networks internal distance.
Projections of how much we are going to get from CIDR will be somewhat soft until we started getting a couple of real life examples. I have poked at what we would get under various CIDR strategeies, but as I have not done it for "real", I am not 100% sure yet.
Here is some results from a run that I did of my data a while back: 1381 total nets 646 CIDR routes if CIDRize by site 619 CIDR routes if CIDRize by AUP (NSFNET vs non-NSFNET) 1026 NSFNET nets; 468 NSFNET CIDR routes 355 non-NSFNET nets; 151 non-NSFNET CIDR routes 500 CIDR routes if CIDRize everything together
Even if we just CIDRized by site, it looks like we will get nearly all of the savings that we would get if we fully CIDRized. --asp@uunet.uu.net (Andrew Partan)
Alternet has grown very substantially since CIDR allocation was in place, much more so than many other networks (gov agencies, some regionals that were well established early on, etc). So Alternet represents a good case to benfit from CIDR, but maybe not a typical case. I'm glad to hear that Alternet alone can eliminate 700-800 of the routes. Things look optimistic for getting relief from the 19,000 route situation we are now in, if we can just get rolling. Curtis
This was done without including holes so b) is the right answer ! --T Jessica Yu <jyy@merit.edu> writes: * * The first table focuses on the current classful routes being passed around * by AS * and the amount of gain CIDR would give *IF* we could aggregate as the AS le * vel. * It shows the number of nets per AS so we see who is announcing what in term * s of * BGP as well as what the number of CIDR nets would be advertised if we could * aggregate at this level... * * Tony, * * Interesting. Just wonder if the would CIDR nets (aggregates) of each A * S * that you composed includes holes or not. E.g. * * If AS1 has routes * 199.1.0 - 199.1.7 except 199.1.3. * What CIDR net does it count for this AS in your list: * * a. 199.1.0/21 (aggregate with hole and reduce routes from 7 to 1) or * b. 199.1.0, 199.1.1, 199.1.2 & 199.1.4/22 (without hole and reduce * routes from 7 to 4) * * Basically, I like to know if this reduction of routes on your list (32. * 6%) * is calculated with aggregates with holes or not. If it does not * include holes, we should be able to see a MORE reduction of the routing * table size by doing the same analysis but include holes in the potentia * l * aggregates. * * --Jessica
participants (2)
-
asp＠uunet.uu.net -
Curtis Villamizar -
Jessica Yu -
Tony Bates