idnits 2.17.1 draft-morton-ippm-initial-registry-01.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- -- The document has examples using IPv4 documentation addresses according to RFC6890, but does not use any IPv6 documentation addresses. Maybe there should be IPv6 examples, too? Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 597 has weird spacing: '... Src the IP...' == Line 600 has weird spacing: '... Dst the IP...' == Line 612 has weird spacing: '.../lambda avera...' == Line 631 has weird spacing: '... Src launch...' == Line 634 has weird spacing: '... Dst waits ...' -- The document date (October 18, 2015) is 3113 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: 'RFC2680' is defined on line 2661, but no explicit reference was found in the text == Unused Reference: 'RFC4737' is defined on line 2680, but no explicit reference was found in the text == Unused Reference: 'Brow00' is defined on line 2705, but no explicit reference was found in the text == Unused Reference: 'RFC4148' is defined on line 2717, but no explicit reference was found in the text == Unused Reference: 'RFC5472' is defined on line 2725, but no explicit reference was found in the text == Unused Reference: 'RFC5477' is defined on line 2730, but no explicit reference was found in the text == Unused Reference: 'RFC6248' is defined on line 2739, but no explicit reference was found in the text == Unused Reference: 'RFC7594' is defined on line 2770, but no explicit reference was found in the text ** Downref: Normative reference to an Informational RFC: RFC 2330 ** Obsolete normative reference: RFC 2679 (Obsoleted by RFC 7679) ** Obsolete normative reference: RFC 2680 (Obsoleted by RFC 7680) -- Obsolete informational reference (is this intentional?): RFC 4148 (Obsoleted by RFC 6248) -- Obsolete informational reference (is this intentional?): RFC 4566 (Obsoleted by RFC 8866) Summary: 3 errors (**), 0 flaws (~~), 14 warnings (==), 4 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group A. Morton 3 Internet-Draft AT&T Labs 4 Intended status: Standards Track M. Bagnulo 5 Expires: April 20, 2016 UC3M 6 P. Eardley 7 BT 8 K. D'Souza 9 AT&T Labs 10 October 18, 2015 12 Initial Performance Metric Registry Entries 13 draft-morton-ippm-initial-registry-01 15 Abstract 17 This memo defines the Initial Entries for the Performance Metrics 18 Registry. 20 Section 4 has been edited to reflect recent discussion on the ippm- 21 list: * Removed the combination or "Raw" and left 95th percentile. * 22 Hanging Indent on Run-time parameters (Fixed parameters use bullet 23 lists and other indenting formats. * Payload format for measurement 24 has been removed. * Explanation of Conditional delay distribution. 26 Still need: * suggestion of standard naming format for parameters. * 27 usecase(s) for machine parse-able registry columns. * Contribution of 28 YANG references for alternate data formats. 30 Note: lambda parameter description is correct in section 4, elsewhere 31 needs fix. 33 Requirements Language 35 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 36 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 37 document are to be interpreted as described in RFC 2119 [RFC2119]. 39 Status of This Memo 41 This Internet-Draft is submitted in full conformance with the 42 provisions of BCP 78 and BCP 79. 44 Internet-Drafts are working documents of the Internet Engineering 45 Task Force (IETF). Note that other groups may also distribute 46 working documents as Internet-Drafts. The list of current Internet- 47 Drafts is at http://datatracker.ietf.org/drafts/current/. 49 Internet-Drafts are draft documents valid for a maximum of six months 50 and may be updated, replaced, or obsoleted by other documents at any 51 time. It is inappropriate to use Internet-Drafts as reference 52 material or to cite them other than as "work in progress." 54 This Internet-Draft will expire on April 20, 2016. 56 Copyright Notice 58 Copyright (c) 2015 IETF Trust and the persons identified as the 59 document authors. All rights reserved. 61 This document is subject to BCP 78 and the IETF Trust's Legal 62 Provisions Relating to IETF Documents 63 (http://trustee.ietf.org/license-info) in effect on the date of 64 publication of this document. Please review these documents 65 carefully, as they describe your rights and restrictions with respect 66 to this document. Code Components extracted from this document must 67 include Simplified BSD License text as described in Section 4.e of 68 the Trust Legal Provisions and are provided without warranty as 69 described in the Simplified BSD License. 71 Table of Contents 73 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 7 74 2. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 75 3. Registry Categories and Columns . . . . . . . . . . . . . . . 8 76 4. UDP Round-trip Latency Registry Entry . . . . . . . . . . . . 9 77 4.1. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 10 78 4.1.1. ID (Identifier) . . . . . . . . . . . . . . . . . . . 10 79 4.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 10 80 4.1.3. URIs . . . . . . . . . . . . . . . . . . . . . . . . 10 81 4.1.4. Description . . . . . . . . . . . . . . . . . . . . . 10 82 4.2. Metric Definition . . . . . . . . . . . . . . . . . . . . 10 83 4.2.1. Reference Definition . . . . . . . . . . . . . . . . 10 84 4.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 11 85 4.3. Method of Measurement . . . . . . . . . . . . . . . . . . 12 86 4.3.1. Reference Method . . . . . . . . . . . . . . . . . . 12 87 4.3.2. Packet Generation Stream . . . . . . . . . . . . . . 13 88 4.3.3. Traffic Filtering (observation) Details . . . . . . . 13 89 4.3.4. Sampling Distribution . . . . . . . . . . . . . . . . 13 90 4.3.5. Run-time Parameters and Data Format . . . . . . . . . 14 91 4.3.6. Roles . . . . . . . . . . . . . . . . . . . . . . . . 14 92 4.4. Output . . . . . . . . . . . . . . . . . . . . . . . . . 15 93 4.4.1. Type . . . . . . . . . . . . . . . . . . . . . . . . 15 94 4.4.2. Data Format . . . . . . . . . . . . . . . . . . . . . 15 95 4.4.3. Reference . . . . . . . . . . . . . . . . . . . . . . 16 96 4.4.4. Metric Units . . . . . . . . . . . . . . . . . . . . 16 98 4.5. Administrative items . . . . . . . . . . . . . . . . . . 16 99 4.5.1. Status . . . . . . . . . . . . . . . . . . . . . . . 16 100 4.5.2. Requestor (keep?) . . . . . . . . . . . . . . . . . . 16 101 4.5.3. Revision . . . . . . . . . . . . . . . . . . . . . . 16 102 4.5.4. Revision Date . . . . . . . . . . . . . . . . . . . . 16 103 4.6. Comments and Remarks . . . . . . . . . . . . . . . . . . 16 104 5. Packet Delay Variation Registry Entry . . . . . . . . . . . . 16 105 5.1. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 17 106 5.1.1. ID (Identifier) . . . . . . . . . . . . . . . . . . . 17 107 5.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 17 108 5.1.3. URI . . . . . . . . . . . . . . . . . . . . . . . . . 17 109 5.1.4. Description . . . . . . . . . . . . . . . . . . . . . 17 110 5.2. Metric Definition . . . . . . . . . . . . . . . . . . . . 17 111 5.2.1. Reference Definition . . . . . . . . . . . . . . . . 17 112 5.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 18 113 5.3. Method of Measurement . . . . . . . . . . . . . . . . . . 18 114 5.3.1. Reference Method . . . . . . . . . . . . . . . . . . 18 115 5.3.2. Packet Generation Stream . . . . . . . . . . . . . . 18 116 5.3.3. Traffic Filtering (observation) Details . . . . . . . 19 117 5.3.4. Sampling Distribution . . . . . . . . . . . . . . . . 19 118 5.3.5. Run-time Parameters and Data Format . . . . . . . . . 19 119 5.3.6. Roles . . . . . . . . . . . . . . . . . . . . . . . . 19 120 5.4. Output . . . . . . . . . . . . . . . . . . . . . . . . . 20 121 5.4.1. Type/Value (two diff terms used) . . . . . . . . . . 20 122 5.4.2. Data Format . . . . . . . . . . . . . . . . . . . . . 20 123 5.4.3. Reference . . . . . . . . . . . . . . . . . . . . . . 21 124 5.4.4. Metric Units . . . . . . . . . . . . . . . . . . . . 21 125 5.5. Administrative items . . . . . . . . . . . . . . . . . . 21 126 5.5.1. Status . . . . . . . . . . . . . . . . . . . . . . . 21 127 5.5.2. Requestor (keep?) . . . . . . . . . . . . . . . . . . 22 128 5.5.3. Revision . . . . . . . . . . . . . . . . . . . . . . 22 129 5.5.4. Revision Date . . . . . . . . . . . . . . . . . . . . 22 130 5.6. Comments and Remarks . . . . . . . . . . . . . . . . . . 22 131 6. DNS Response Latency Registry Entry . . . . . . . . . . . . . 22 132 6.1. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 22 133 6.1.1. ID (Identifier) . . . . . . . . . . . . . . . . . . . 22 134 6.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 22 135 6.1.3. URI . . . . . . . . . . . . . . . . . . . . . . . . . 23 136 6.1.4. Description . . . . . . . . . . . . . . . . . . . . . 23 137 6.2. Metric Definition . . . . . . . . . . . . . . . . . . . . 23 138 6.2.1. Reference Definition . . . . . . . . . . . . . . . . 23 139 6.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 24 140 6.3. Method of Measurement . . . . . . . . . . . . . . . . . . 25 141 6.3.1. Reference Method . . . . . . . . . . . . . . . . . . 25 142 6.3.2. Packet Generation Stream . . . . . . . . . . . . . . 26 143 6.3.3. Traffic Filtering (observation) Details . . . . . . . 26 144 6.3.4. Sampling Distribution . . . . . . . . . . . . . . . . 26 145 6.3.5. Run-time Parameters and Data Format . . . . . . . . . 27 146 6.3.6. Roles . . . . . . . . . . . . . . . . . . . . . . . . 28 147 6.4. Output . . . . . . . . . . . . . . . . . . . . . . . . . 28 148 6.4.1. Type/Value (two diff terms used) . . . . . . . . . . 28 149 6.4.2. Data Format . . . . . . . . . . . . . . . . . . . . . 28 150 6.4.3. Reference . . . . . . . . . . . . . . . . . . . . . . 29 151 6.4.4. Metric Units . . . . . . . . . . . . . . . . . . . . 29 152 6.5. Administrative items . . . . . . . . . . . . . . . . . . 29 153 6.5.1. Status . . . . . . . . . . . . . . . . . . . . . . . 29 154 6.5.2. Requestor (keep?) . . . . . . . . . . . . . . . . . . 30 155 6.5.3. Revision . . . . . . . . . . . . . . . . . . . . . . 30 156 6.5.4. Revision Date . . . . . . . . . . . . . . . . . . . . 30 157 6.6. Comments and Remarks . . . . . . . . . . . . . . . . . . 30 158 7. UDP Poisson One-way Delay Registry Entries . . . . . . . . . 30 159 7.1. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 30 160 7.1.1. ID (Identifier) . . . . . . . . . . . . . . . . . . . 30 161 7.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 30 162 7.1.3. URI and URL . . . . . . . . . . . . . . . . . . . . . 31 163 7.1.4. Description . . . . . . . . . . . . . . . . . . . . . 31 164 7.2. Metric Definition . . . . . . . . . . . . . . . . . . . . 31 165 7.2.1. Reference Definition . . . . . . . . . . . . . . . . 31 166 7.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 32 167 7.3. Method of Measurement . . . . . . . . . . . . . . . . . . 32 168 7.3.1. Reference Method . . . . . . . . . . . . . . . . . . 32 169 7.3.2. Packet Generation Stream . . . . . . . . . . . . . . 33 170 7.3.3. Traffic Filtering (observation) Details . . . . . . . 33 171 7.3.4. Sampling Distribution . . . . . . . . . . . . . . . . 33 172 7.3.5. Run-time Parameters and Data Format . . . . . . . . . 33 173 7.3.6. Roles . . . . . . . . . . . . . . . . . . . . . . . . 34 174 7.4. Output . . . . . . . . . . . . . . . . . . . . . . . . . 34 175 7.4.1. Type/Value (two diff terms used) . . . . . . . . . . 34 176 7.4.2. Data Format . . . . . . . . . . . . . . . . . . . . . 35 177 7.4.3. Reference . . . . . . . . . . . . . . . . . . . . . . 37 178 7.4.4. Metric Units . . . . . . . . . . . . . . . . . . . . 37 179 7.5. Administrative items . . . . . . . . . . . . . . . . . . 37 180 7.5.1. Status . . . . . . . . . . . . . . . . . . . . . . . 37 181 7.5.2. Requestor (keep?) . . . . . . . . . . . . . . . . . . 37 182 7.5.3. Revision . . . . . . . . . . . . . . . . . . . . . . 37 183 7.5.4. Revision Date . . . . . . . . . . . . . . . . . . . . 37 184 7.6. Comments and Remarks . . . . . . . . . . . . . . . . . . 37 185 8. UDP Periodic One-way Delay Registry Entries . . . . . . . . . 37 186 8.1. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 38 187 8.1.1. ID (Identifier) . . . . . . . . . . . . . . . . . . . 38 188 8.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 38 189 8.1.3. URI and URL . . . . . . . . . . . . . . . . . . . . . 38 190 8.1.4. Description . . . . . . . . . . . . . . . . . . . . . 38 191 8.2. Metric Definition . . . . . . . . . . . . . . . . . . . . 39 192 8.2.1. Reference Definition . . . . . . . . . . . . . . . . 39 193 8.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 39 195 8.3. Method of Measurement . . . . . . . . . . . . . . . . . . 40 196 8.3.1. Reference Method . . . . . . . . . . . . . . . . . . 40 197 8.3.2. Packet Generation Stream . . . . . . . . . . . . . . 40 198 8.3.3. Traffic Filtering (observation) Details . . . . . . . 41 199 8.3.4. Sampling Distribution . . . . . . . . . . . . . . . . 41 200 8.3.5. Run-time Parameters and Data Format . . . . . . . . . 41 201 8.3.6. Roles . . . . . . . . . . . . . . . . . . . . . . . . 42 202 8.4. Output . . . . . . . . . . . . . . . . . . . . . . . . . 42 203 8.4.1. Type/Value (two diff terms used) . . . . . . . . . . 42 204 8.4.2. Data Format . . . . . . . . . . . . . . . . . . . . . 42 205 8.4.3. Reference . . . . . . . . . . . . . . . . . . . . . . 44 206 8.4.4. Metric Units . . . . . . . . . . . . . . . . . . . . 44 207 8.5. Administrative items . . . . . . . . . . . . . . . . . . 45 208 8.5.1. Status . . . . . . . . . . . . . . . . . . . . . . . 45 209 8.5.2. Requestor (keep?) . . . . . . . . . . . . . . . . . . 45 210 8.5.3. Revision . . . . . . . . . . . . . . . . . . . . . . 45 211 8.5.4. Revision Date . . . . . . . . . . . . . . . . . . . . 45 212 8.6. Comments and Remarks . . . . . . . . . . . . . . . . . . 45 213 9. partly BLANK Registry Entry . . . . . . . . . . . . . . . . . 45 214 9.1. Summary . . . . . . . . . . . . . . . . . . . . . . . . . 45 215 9.1.1. ID (Identifier) . . . . . . . . . . . . . . . . . . . 45 216 9.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 45 217 9.1.3. URI . . . . . . . . . . . . . . . . . . . . . . . . . 45 218 9.1.4. Description . . . . . . . . . . . . . . . . . . . . . 46 219 9.2. Metric Definition . . . . . . . . . . . . . . . . . . . . 46 220 9.2.1. Reference Definition . . . . . . . . . . . . . . . . 46 221 9.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 46 222 9.3. Method of Measurement . . . . . . . . . . . . . . . . . . 47 223 9.3.1. Reference Method . . . . . . . . . . . . . . . . . . 47 224 9.3.2. Packet Generation Stream . . . . . . . . . . . . . . 47 225 9.3.3. Traffic Filtering (observation) Details . . . . . . . 47 226 9.3.4. Sampling Distribution . . . . . . . . . . . . . . . . 48 227 9.3.5. Run-time Parameters and Data Format . . . . . . . . . 48 228 9.3.6. Roles . . . . . . . . . . . . . . . . . . . . . . . . 48 229 9.4. Output . . . . . . . . . . . . . . . . . . . . . . . . . 48 230 9.4.1. Type/Value (two diff terms used) . . . . . . . . . . 48 231 9.4.2. Data Format . . . . . . . . . . . . . . . . . . . . . 48 232 9.4.3. Reference . . . . . . . . . . . . . . . . . . . . . . 48 233 9.4.4. Metric Units . . . . . . . . . . . . . . . . . . . . 49 234 9.5. Administrative items . . . . . . . . . . . . . . . . . . 49 235 9.5.1. Status . . . . . . . . . . . . . . . . . . . . . . . 49 236 9.5.2. Requestor (keep?) . . . . . . . . . . . . . . . . . . 49 237 9.5.3. Revision . . . . . . . . . . . . . . . . . . . . . . 49 238 9.5.4. Revision Date . . . . . . . . . . . . . . . . . . . . 49 239 9.6. Comments and Remarks . . . . . . . . . . . . . . . . . . 49 240 10. BLANK Registry Entry . . . . . . . . . . . . . . . . . . . . 49 241 10.1. Summary . . . . . . . . . . . . . . . . . . . . . . . . 49 242 10.1.1. ID (Identifier) . . . . . . . . . . . . . . . . . . 49 243 10.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 49 244 10.1.3. URI . . . . . . . . . . . . . . . . . . . . . . . . 50 245 10.1.4. Description . . . . . . . . . . . . . . . . . . . . 50 246 10.2. Metric Definition . . . . . . . . . . . . . . . . . . . 50 247 10.2.1. Reference Definition . . . . . . . . . . . . . . . . 50 248 10.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 50 249 10.3. Method of Measurement . . . . . . . . . . . . . . . . . 50 250 10.3.1. Reference Method . . . . . . . . . . . . . . . . . . 50 251 10.3.2. Packet Generation Stream . . . . . . . . . . . . . . 50 252 10.3.3. Traffic Filtering (observation) Details . . . . . . 50 253 10.3.4. Sampling Distribution . . . . . . . . . . . . . . . 51 254 10.3.5. Run-time Parameters and Data Format . . . . . . . . 51 255 10.3.6. Roles . . . . . . . . . . . . . . . . . . . . . . . 51 256 10.4. Output . . . . . . . . . . . . . . . . . . . . . . . . . 51 257 10.4.1. Type/Value (two diff terms used) . . . . . . . . . . 51 258 10.4.2. Data Format . . . . . . . . . . . . . . . . . . . . 51 259 10.4.3. Reference . . . . . . . . . . . . . . . . . . . . . 51 260 10.4.4. Metric Units . . . . . . . . . . . . . . . . . . . . 51 261 10.5. Administrative items . . . . . . . . . . . . . . . . . . 51 262 10.5.1. Status . . . . . . . . . . . . . . . . . . . . . . . 51 263 10.5.2. Requestor (keep?) . . . . . . . . . . . . . . . . . 51 264 10.5.3. Revision . . . . . . . . . . . . . . . . . . . . . . 52 265 10.5.4. Revision Date . . . . . . . . . . . . . . . . . . . 52 266 10.6. Comments and Remarks . . . . . . . . . . . . . . . . . . 52 267 11. Example RTCP-XR Registry Entry . . . . . . . . . . . . . . . 52 268 11.1. Registry Indexes . . . . . . . . . . . . . . . . . . . . 52 269 11.1.1. Identifier . . . . . . . . . . . . . . . . . . . . . 52 270 11.1.2. Name . . . . . . . . . . . . . . . . . . . . . . . . 52 271 11.1.3. URI . . . . . . . . . . . . . . . . . . . . . . . . 52 272 11.1.4. Status . . . . . . . . . . . . . . . . . . . . . . . 52 273 11.1.5. Requestor . . . . . . . . . . . . . . . . . . . . . 52 274 11.1.6. Revision . . . . . . . . . . . . . . . . . . . . . . 53 275 11.1.7. Revision Date . . . . . . . . . . . . . . . . . . . 53 276 11.1.8. Description . . . . . . . . . . . . . . . . . . . . 53 277 11.1.9. Reference Specification(s) . . . . . . . . . . . . . 53 278 11.2. Metric Definition . . . . . . . . . . . . . . . . . . . 53 279 11.2.1. Reference Definition . . . . . . . . . . . . . . . . 53 280 11.2.2. Fixed Parameters . . . . . . . . . . . . . . . . . . 53 281 11.3. Method of Measurement . . . . . . . . . . . . . . . . . 54 282 11.3.1. Reference Method . . . . . . . . . . . . . . . . . . 54 283 11.3.2. Stream Type and Stream Parameters . . . . . . . . . 54 284 11.3.3. Output Type and Data Format . . . . . . . . . . . . 55 285 11.3.4. Metric Units . . . . . . . . . . . . . . . . . . . . 55 286 11.3.5. Run-time Parameters and Data Format . . . . . . . . 55 287 11.4. Comments and Remarks . . . . . . . . . . . . . . . . . . 57 288 12. Security Considerations . . . . . . . . . . . . . . . . . . . 57 289 13. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 57 290 14. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 57 291 15. References . . . . . . . . . . . . . . . . . . . . . . . . . 57 292 15.1. Normative References . . . . . . . . . . . . . . . . . . 57 293 15.2. Informative References . . . . . . . . . . . . . . . . . 59 294 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 60 296 1. Introduction 298 Note: Efforts to synchronize structure and terminology with 299 [I-D.ietf-ippm-metric-registry] will likely be incomplete until both 300 drafts are stable. 302 This memo defines the Initial set of entries for the Performance 303 Metric Registry. The registry will contain Active Performance 304 Metrics, especially those defined in RFCs prepared in the IP 305 Performance Metrics (IPPM) Working Group of the IETF, according to 306 their framework [RFC2330]. Three aspects make IPPM metric 307 registration difficult: (1) Use of the Type-P notion to allow users 308 to specify their own packet types. (2) Use of Flexible input 309 variables, called Parameters in IPPM definitions, some which 310 determine the quantity measured and others which should not be 311 specified until execution of the measurement. (3) Allowing 312 flexibility in choice of statistics to summarize the results on a 313 stream of measurement packets. This memo uses terms and definitions 314 from the IPPM literature, primarily [RFC2330], and the reader is 315 assumed familiar with them or may refer questions there as necessary. 317 Although there are several standard templates for organizing 318 specifications of performance metrics (see [RFC2679] for an example 319 of the traditional IPPM template, based to large extent on the 320 Benchmarking Methodology Working Group's traditional template in 321 [RFC1242], and see [RFC6390] for a similar template), none of these 322 templates were intended to become the basis for the columns of an 323 IETF-wide registry of metrics. While examinating aspects of metric 324 specifications which need to be registered, it was clear that none of 325 the existing metric templates fully satisfies the particular needs of 326 a registry. 328 2. Scope 330 [I-D.ietf-ippm-metric-registry] defines the overall structure for a 331 Performance Metric Registry and provides guidance for the process to 332 examine proposed metrics and maitain Registered Metrics. 334 This document defines the initial set of Performance Metrics Registry 335 entries; all are active metrics, or those where the packets measured 336 have been specially generated for the purpose of measurement. 338 A row in the registry corresponds to one Registered Performance 339 Metric, with entries in the various columns specifying the metric. 341 As discussed in [I-D.ietf-ippm-metric-registry], each entry (row) 342 must be tightly defined; the definition must leave open only a few 343 parameters that do not change the fundamental nature of the 344 measurement (such as source and destination addresses), and so 345 promotes comparable results across independent implementations. 346 Also, each registered entry must be based on existing reference RFCs 347 (or other standards) for performance metrics, and must be 348 operationally useful and have significant industry interest. This is 349 ensured by expert review for every entry before IANA action. 351 3. Registry Categories and Columns 353 This section defines the categories and columns of the registry. 354 Below, categories are described at the 3.x heading level, and columns 355 are at the 3.x.y heading level. The Figure below illustrates this 356 organization. An entry (row) therefore gives a complete description 357 of a Registered Metric. 359 Each column serves as a check-list item and helps to avoid omissions 360 during registration and expert review. In some cases an entry (row) 361 may have some columns without specific entries, marked Not Applicable 362 (NA). 364 Registry Categories and Columns, shown as 365 Category 366 ------------------ 367 Column | Column | 369 Summary 370 -------------------------------- 371 ID | Name | URIs | Description | 373 Metric Definition 374 ----------------------------------------- 375 Reference Definition | Fixed Parameters | 377 Method of Measurement 378 --------------------------------------------------------------- 379 Reference | Packet | Traffic | Sampling | Run-time | Role | 380 Method | Generation | Filter | dist. | Param | | 381 | Stream | 383 Output 384 ---------------------------- 385 Type | Reference | Units | 386 | Definition | | 388 Administrative information 389 ---------------------------------- 390 Status |Request | Rev | Rev.Date | 392 Comments and Remarks 393 -------------------- 395 4. UDP Round-trip Latency Registry Entry 397 This section gives an initial registry entry for the UDP Round-trip 398 Latency. 400 Note: If each Registry entry should only produce a "raw" output or a 401 statistical summary, then the "Output" Category can be split and this 402 section can become two closely-related metrics. 404 4.1. Summary 406 This category includes multiple indexes to the registry entries, the 407 element ID and metric name. 409 4.1.1. ID (Identifier) 411 413 4.1.2. Name 415 417 Act_IP_UDP_Round-trip_Delay_95th-percentile_Poisson 419 4.1.3. URIs 421 URN: Prefix urn:ietf:params:performance:metric... 423 URL: http:/// 425 4.1.4. Description 427 This metric assesses the delay of a stream of packets exchanged 428 between two hosts (or measurement points), and the Output is the 429 Round-trip delay for all successfully exchanged packets expressed as 430 the 95th percentile of their conditional delay distribution. 432 4.2. Metric Definition 434 This category includes columns to prompt the entry of all necessary 435 details related to the metric definition, including the RFC reference 436 and values of input factors, called fixed parameters. 438 4.2.1. Reference Definition 440 442 Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip Delay 443 Metric for IPPM", RFC 2681, September 1999. 445 [RFC2681] 447 449 Section 2.4 of [RFC2681] provides the reference definition of the 450 singleton (single value) Round-trip delay metric. Section 3.4 of 451 [RFC2681] provides the reference definition expanded to cover a 452 multi-value sample. Note that terms such as singleton and sample are 453 defined in Section 11 of [RFC2330]. 455 Note that although the definition of "Round-trip-Delay between Src 456 and Dst at T" is directionally ambiguous in the text, this metric 457 tightens the definition further to recognize that the host in the 458 "Src" role will send the first packet to "Dst", and ultimately 459 receive the corresponding return packet from "Dst" (when neither are 460 lost). 462 Finally, note that the variable "dT" is used in [RFC2681] to refer to 463 the value of Round-trip delay in metric definitions and methods. The 464 variable "dT" has been re-used in other IPPM literature to refer to 465 different quantities, and cannot be used as a global variable name. 467 4.2.2. Fixed Parameters 469 473 Type-P: 475 o IPv4 header values: 477 * DSCP: set to 0 479 * TTL: set to 255 481 * Protocol: Set to 17 (UDP) 483 o UDP header values: 485 * Checksum: the checksum must be calculated 487 o UDP Payload 489 * total of 9 bytes 491 Tmax: a loss threshold waiting time 493 o 3 seconds, expressed as 32-bit NTP timestamp as per section 6 of 494 RFC 5905 [RFC5905] 496 4.3. Method of Measurement 498 This category includes columns for references to relevant sections of 499 the RFC(s) and any supplemental information needed to ensure an 500 unambiguous methods for implementations. 502 4.3.1. Reference Method 504 507 The methodology for this metric is defined as Type-P-Round-trip- 508 Delay-Poisson-Stream in section 2.6 of RFC 2681 [RFC2681] and section 509 3.6 of RFC 2681 [RFC2681] using the Type-P and Tmax defined under 510 Fixed Parameters. 512 The reference method distinguishes between long-delayed packets and 513 lost packets by implementing a maximum waiting time for packet 514 arrival. Tmax is the waiting time used as the threshold to declare a 515 packet lost. Lost packets SHALL be designated as having undefined 516 delay. 518 The calculations on the delay (RTT) SHALL be performed on the 519 conditional distribution, conditioned on successful packet arrival 520 within Tmax. Also, when all packet delays are stored, the process 521 which calculates the RTT value MAY enforce the Tmax threshold on 522 stored values before calculations. See section 4.1 of [RFC3393] for 523 details on the conditional distribution to exclude undefined values 524 of delay, and Section 5 of [RFC6703] for background on this analysis 525 choice. 527 The reference method requires some way to distinguish between 528 different packets in a stream and establish correspondence between 529 sending times and receiving times for each successfully-arriving 530 packet. Sequence numbers or other send-order information MUST be 531 retained at the Src or included with each packet to dis-ambiguate 532 packet reordering if it occurs. The chosen measurement protocol will 533 dictate the format of sequence numbers and time-stamps, if they are 534 conveyed in the packet payload. The sequence numbers or timestamps 535 MUST be determined by the measurement process itself after the Fixed 536 and Runtime parameters are passed to that process. 538 Refer to Section 4.4 of [RFC6673] for expanded discussion of the 539 instruction to "send a Type-P packet back to the Src as quickly as 540 possible" in Section 2.6 of RFC 2681 [RFC2681]. Section 8 of 541 [RFC6673] presents additional requirements which shall be included in 542 the method of measurement for this metric. 544 4.3.2. Packet Generation Stream 546 This section gives the details of the packet traffic which is the 547 basis for measurement. In IPPM metrics, this is called the Stream, 548 and can easily be described by providing the list of stream 549 parameters. 551
554 Section 11.1.3 of [RFC2330] provides three methods to generate 555 Poisson sampling intervals. the reciprocal of lambda is the average 556 packet spacing, thus the Run-time Parameter is 1/lambda, (in 557 seconds). 559 >>> Check with Sam, most likely it is this... 561 Method 3 SHALL be used, where given a start time (Run-time 562 Parameter), the subsequent send times are all computed prior to 563 measurement by computing the pseudo-random distribution of inter- 564 packet send times, (truncating the distribution as specified in the 565 Run-time Parameter, Trunc), and the Src sends each packet at the 566 computed times. 568 Note that Trunc is the upper limit on inter-packet times in the 569 Poisson distribution. A random value greater than Trunc is set equal 570 to Trunc instead. 572 4.3.3. Traffic Filtering (observation) Details 574 The measured results based on a filtered version of the packets 575 observed, and this section provides the filter details (when 576 present). 578
. 580 NA 582 4.3.4. Sampling Distribution 584 587 NA 589 4.3.5. Run-time Parameters and Data Format 591 Run-time Parameters are input factors that must be determined, 592 configured into the measurement system, and reported with the results 593 for the context to be complete. 595 597 Src the IP address of a host (32-bit value for IPv4, 128-bit value 598 for IPv6) 600 Dst the IP address of a host (32-bit value for IPv4, 128-bit value 601 for IPv6) 603 T0 a time (start of measurement interval, 128-bit NTP Date Format, 604 see section 6 of [RFC5905]). When T0 is "all-zeros", a start time 605 is unspecified and Tf is to be interpreted as the Duration of the 606 measurement interval. 608 Tf a time (end of measurement interval, 128-bit NTP Date Format, see 609 section 6 of [RFC5905]), interpreted as the Duration of the 610 measurement interval. 612 1/lambda average packet interval (for Poisson Streams). (1/lambda = 613 1 second, if fixed) 615 Trunc Upper limit on Poisson distribution (values above this limit 616 will be clipped and set to the limit value). (if fixed, Trunc = 617 30 seconds.) 619 The format for 1/lambda and Trunc of Poisson Dist. are the short 620 format in [RFC5905] (32 bits) and is as follows: the first 16 bits 621 represent the integer number of seconds; the next 16 bits represent 622 the fractional part of a second. 624 >>> should Poisson run-time params be fixed instead? probably yes if 625 modeling a specific version of MBA tests. 627 4.3.6. Roles 629 631 Src launches each packet and waits for return transmissions from 632 Dst. 634 Dst waits for each packet from Src and sends a return packet to Src. 636 4.4. Output 638 This category specifies all details of the Output of measurements 639 using the metric. 641 4.4.1. Type 643 645 Percentile -- for the conditional distribution of all packets with a 646 valid value of Round-trip delay (undefined delays are excluded), a 647 single value corresponding to the 95th percentile. 649 4.4.2. Data Format 651 653 For all outputs --- 655 T0 a time (start of measurement interval, 128-bit NTP Date Format, 656 see section 6 of [RFC5905]) 658 Tf a time (end of measurement interval, 128-bit NTP Date Format, see 659 section 6 of [RFC5905]) 661 Raw -- REMOVED IN VERSION 01 663 Percentile -- for the conditional distribution of all packets with a 664 valid value of Round-trip delay (undefined delays are excluded), a 665 single value as follows: 667 See section 4.1 of [RFC3393] for details on the conditional 668 distribution to exclude undefined values of delay, and Section 5 of 669 [RFC6703] for background on this analysis choice. 671 See section 4.3 of [RFC3393] for details on the percentile statistic 672 (where Round-trip delay should be substituted for "ipdv"). 674 The percentile = 95. 676 Data format is a 32-bit signed value, *similar to* the 32-bit short 677 NTP Time format in Section 6 of [RFC5905] and is as follows: the 678 first 16 bits represent the *signed* integer number of seconds; the 679 next 16 bits represent the fractional part of a second. 681 4.4.3. Reference 683 685 See the Data Format column for references. 687 4.4.4. Metric Units 689 . 692 The 95th Percentile of Round-trip Delay is expressed in seconds. 694 4.5. Administrative items 696 4.5.1. Status 698 700 4.5.2. Requestor (keep?) 702 name or RFC, etc. 704 4.5.3. Revision 706 1.0 708 4.5.4. Revision Date 710 YYYY-MM-DD 712 4.6. Comments and Remarks 714 Additional (Informational) details for this entry 716 5. Packet Delay Variation Registry Entry 718 This section gives an initial registry entry for a Packet Delay 719 Variation metric. 721 Note: If each Registry entry should only produce a "raw" output or a 722 statistical summary, then the "Output" Category can be split and this 723 section can become two closely-related metrics. 725 5.1. Summary 727 This category includes multiple indexes to the registry entries, the 728 element ID and metric name. 730 732 5.1.1. ID (Identifier) 734 736 5.1.2. Name 738 740 Act_IP-UDP-One-way-pdv-95th-percentile-Poisson 742 URL: ?? 744 5.1.3. URI 746 URI: Prefix urn:ietf:params:performance:metric 748 5.1.4. Description 750 An assessment of packet delay variation with respect to the minimum 751 delay observed on the stream. 753 5.2. Metric Definition 755 This category includes columns to prompt the entry of all necessary 756 details related to the metric definition, including the RFC reference 757 and values of input factors, called fixed parameters. 759 5.2.1. Reference Definition 761 763 Paxson, V., Almes, G., Mahdavi, J., and M. Mathis, "Framework for IP 764 Performance Metrics", RFC 2330, May 1998. [RFC2330] 766 Demichelis, C. and P. Chimento, "IP Packet Delay Variation Metric 767 for IP Performance Metrics (IPPM)", RFC 3393, November 2002. 768 [RFC3393] 770 Morton, A. and B. Claise, "Packet Delay Variation Applicability 771 Statement", RFC 5481, March 2009. [RFC5481] 772 Mills, D., Martin, J., Burbank, J., and W. Kasch, "Network Time 773 Protocol Version 4: Protocol and Algorithms Specification", RFC 5905, 774 June 2010.[RFC5905] 776 778 See sections 2.4 and 3.4 of [RFC3393]. Singleton delay differences 779 measured are referred to by the variable name "ddT". 781 5.2.2. Fixed Parameters 783 787 o F, a selection function defining unambiguously the packets from 788 the stream selected for the metric. See section 4.2 of [RFC5481] 789 for the PDV form. 791 o L, a packet length in bits. L = 200 bits. 793 o Tmax, a maximum waiting time for packets to arrive at Dst, set 794 sufficiently long to disambiguate packets with long delays from 795 packets that are discarded (lost). Tmax = 3 seconds. 797 o Type-P, as defined in [RFC2330], which includes any field that may 798 affect a packet's treatment as it traverses the network. The 799 packets are IP/UDP, with DSCP = 0 (BE). 801 5.3. Method of Measurement 803 This category includes columns for references to relevant sections of 804 the RFC(s) and any supplemental information needed to ensure an 805 unambiguous methods for implementations. 807 5.3.1. Reference Method 809 812 See section 2.6 and 3.6 of [RFC3393] for singleton elements. 814 5.3.2. Packet Generation Stream 816 818 Poisson distributed as described in [RFC2330], with the following 819 Parameters. 821 o lambda, a rate in reciprocal seconds (for Poisson Streams). 822 lambda = 1 packet per second 824 o Upper limit on Poisson distribution (values above this limit will 825 be clipped and set to the limit value). Upper limit = 30 seconds. 827 5.3.3. Traffic Filtering (observation) Details 829 . 833 NA 835 5.3.4. Sampling Distribution 837 840 NA 842 5.3.5. Run-time Parameters and Data Format 844 . 846 o Src, the IP address of a host (32-bit value for IPv4, 128-bit 847 value for IPv6) 849 o Dst, the IP address of a host (32-bit value for IPv4, 128-bit 850 value for IPv6) 852 o T, a time (start of measurement interval, 128-bit NTP Date Format, 853 see section 6 of [RFC5905]). When T0 is "all-zeros", a start time 854 is unspecified and Tf is to be interpreted as the Duration of the 855 measurement interval. 857 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 858 see section 6 of [RFC5905]), interpreted as the Duration of the 859 measurement interval. 861 5.3.6. Roles 863 865 Src - the host that sends the stream of packets. 867 Dst - the host that receives the stream of packets. 869 5.4. Output 871 This category specifies all details of the Output of measurements 872 using the metric. 874 5.4.1. Type/Value (two diff terms used) 876 878 Raw -- for each packet sent, pairs of values. 880 Percentile -- for the conditional distribution of all packets with a 881 valid value of one-way delay (undefined delays are excluded), a 882 single value corresponding to the 95th percentile of the singletons, 883 ddT. 885 5.4.2. Data Format 887 889 For all Output types 891 o T, a time (start of measurement interval, 128-bit NTP Date Format, 892 see section 6 of [RFC5905]) 894 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 895 see section 6 of [RFC5905]) 897 Raw - 899 o T1, the wire time of the first packet in a pair, measured at 900 MP(Src) as it leaves for Dst (64-bit NTP Timestamp Format, see 901 section 6 of [RFC5905]). 903 o T2, the wire time of the second packet in a pair, measured at 904 MP(Src) as it leaves for Dst (64-bit NTP Timestamp Format, see 905 section 6 of [RFC5905]). 907 o I(i),I(i+1), i >=0, pairs of times which mark the beginning and 908 ending of the intervals in which the packet stream from which the 909 measurement is taken occurs. Here, I(0) = T0 and assuming that n 910 is the largest index, I(n) = Tf (pairs of 64-bit NTP Timestamp 911 Format, see section 6 of [RFC5905]). 913 o When the one-way delay of a packet in the calculation pair for ddT 914 is undefined, then ddT is undefined for that pair. 916 Percentile -- for the conditional distribution of all packets with a 917 valid value of one-way delay (undefined delays are excluded), a 918 single value as follows: 920 See section 4.1 of [RFC3393] for details on the conditional 921 distribution to exclude undefined values of delay, and Section 5 of 922 [RFC6703] for background on this analysis choice. 924 See section 4.3 of [RFC3393] for details on the percentile statistic 925 (where pdv should be substituted for "ipdv"). 927 The percentile = 95. 929 Data format is a 32-bit signed floating point value, *similar to* the 930 32-bit short NTP Time format in Section 6 of [RFC5905] and is as 931 follows: the first 16 bits represent the *signed* integer number of 932 seconds; the next 16 bits represent the fractional part of a second. 934 5.4.3. Reference 936 938 see Data Format column. 940 5.4.4. Metric Units 942 . 945 See section 3.3 of [RFC3393] for singleton elements, ddT. The units 946 are seconds, and the same units are used for 95th percentile. 948 [RFC2330] recommends that when a time is given, it will be expressed 949 in UTC. 951 The timestamp format (for T, Tf, etc.) is the same as in [RFC5905] 952 (64 bits) and is as follows: the first 32 bits represent the unsigned 953 integer number of seconds elapsed since 0h on 1 January 1900; the 954 next 32 bits represent the fractional part of a second that has 955 elapsed since then. 957 5.5. Administrative items 959 5.5.1. Status 961 963 5.5.2. Requestor (keep?) 965 967 5.5.3. Revision 969 1.0 971 5.5.4. Revision Date 973 YYYY-MM-DD 975 5.6. Comments and Remarks 977 979 Lost packets represent a challenge for delay variation metrics. See 980 section 4.1 of [RFC3393] and the delay variation applicability 981 statement[RFC5481] for extensive analysis and comparison of PDV and 982 an alternate metric, IPDV. 984 6. DNS Response Latency Registry Entry 986 This section gives an initial registry entry for DNS Response 987 Latency. RFC 2681 [RFC2681] defines a Round-trip delay metric. We 988 build on that metric by specifying several of the input parameters to 989 precisely define a metric for measuring DNS latency. 991 6.1. Summary 993 This category includes multiple indexes to the registry entries, the 994 element ID and metric name. 996 998 6.1.1. ID (Identifier) 1000 1002 6.1.2. Name 1004 1006 URL: ?? 1008 6.1.3. URI 1010 URI: Prefix urn:ietf:params:performance:metric 1012 6.1.4. Description 1014 This metric assesses the response time, the interval from the query 1015 transmission to the response. 1017 6.2. Metric Definition 1019 This category includes columns to prompt the entry of all necessary 1020 details related to the metric definition, including the RFC reference 1021 and values of input factors, called fixed parameters. 1023 6.2.1. Reference Definition 1025 1027 Mockapetris, P., "Domain names - implementation and specification", 1028 STD 13, RFC 1035, November 1987. (and updates) 1030 [RFC1035] 1032 Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip Delay 1033 Metric for IPPM", RFC 2681, September 1999. 1035 [RFC2681] 1037 1039 Section 2.4 of [RFC2681] provides the reference definition of the 1040 singleton (single value) Round-trip delay metric. Section 3.4 of 1041 [RFC2681] provides the reference definition expanded to cover a 1042 multi-value sample. Note that terms such as singleton and sample are 1043 defined in Section 11 of [RFC2330]. 1045 For DNS Response Latency, the entities in [RFC1035] must be mapped to 1046 [RFC2681]. The Local Host with its User Program and Resolver take 1047 the role of "Src", and the Foreign Name Server takes the role of 1048 "Dst". 1050 Note that although the definition of "Round-trip-Delay between Src 1051 and Dst at T" is directionally ambiguous in the text, this metric 1052 tightens the definition further to recognize that the host in the 1053 "Src" role will send the first packet to "Dst", and ultimately 1054 receive the corresponding return packet from "Dst" (when neither are 1055 lost). 1057 6.2.2. Fixed Parameters 1059 1063 Type-P: 1065 o IPv4 header values: 1067 * DSCP: set to 0 1069 * TTL set to 255 1071 * Protocol: Set to 17 (UDP) 1073 o UDP header values: 1075 * Source port: 53 1077 * Destination port: 53 1079 * Checksum: the checksum must be calculated 1081 o Payload: The payload contains a DNS message as defined in RFC 1035 1082 [RFC1035] with the following values: 1084 * The DNS header section contains: 1086 + QR: set to 0 (Query) 1088 + OPCODE: set to 0 (standard query) 1090 + AA: not set 1092 + TC: not set 1094 + RD: set to one (recursion desired) 1096 + RA: not set 1098 + RCODE: not set 1100 + QDCOUNT: set to one (only one entry) 1102 + ANCOUNT: not set 1104 + NSCOUNT: not set 1105 + ARCOUNT: not set 1107 * The Question section contains: 1109 + QNAME: the FQDN provided as input for the test 1111 + QTYPE: the query type provided as input for the test 1113 + QCLASS: set to IN 1115 * The other sections do not contain any Resource Records. 1117 Observation: reply packets will contain a DNS response and may 1118 contain RRs. 1120 Timeout: Tmax = 5 seconds (to help disambiguate queries) 1122 6.3. Method of Measurement 1124 This category includes columns for references to relevant sections of 1125 the RFC(s) and any supplemental information needed to ensure an 1126 unambiguous methods for implementations. 1128 6.3.1. Reference Method 1130 1133 The methodology for this metric is defined as Type-P-Round-trip- 1134 Delay-Poisson-Stream in section 2.6 of RFC 2681 [RFC2681] and section 1135 3.6 of RFC 2681 [RFC2681] using the Type-P and Timeout defined under 1136 Fixed Parameters. 1138 The method requires sequence numbers or other send-order information 1139 to be retained at the Src or included with each packet to dis- 1140 ambiguate packet reordering if it occurs. Sequence number is part of 1141 the payload described under Fixed Parameters. 1143 DNS Messages bearing Queries provide for random ID Numbers, so more 1144 than one query may be launched while a previous request is 1145 outstanding when the ID Number is used. 1147 IF a DNS response does not arrive within Tmax, the result is 1148 undefined. The Message ID SHALL be used to disambiguate the 1149 successive queries. 1151 >>> This would require support of ID generation and population in the 1152 Message. An alternative would be to use a random Source port on the 1153 Query Message, but we would choose ONE before proceding. 1155 Refer to Section 4.4 of [RFC6673] for expanded discussion of the 1156 instruction to "send a Type-P packet back to the Src as quickly as 1157 possible" in Section 2.6 of RFC 2681 [RFC2681]. Section 8 of 1158 [RFC6673] presents additional requirements which shall be included in 1159 the method of measurement for this metric. 1161 6.3.2. Packet Generation Stream 1163 This section gives the details of the packet traffic which is the 1164 basis for measurement. In IPPM metrics, this is called the Stream, 1165 and can easily be dscribed by providing the list of stream 1166 parameters. 1168 1170 Section 11.1.3 of RFC 2681 [RFC2330] provides three methods to 1171 generate Poisson sampling intervals. the reciprocal of lambda is the 1172 average packet rate, thus the Run-time Parameter is 1/lambda. 1174 >>> Check with Sam, most likely it is this... 1176 Method 3 is used, where given a start time (Run-time Parameter), the 1177 subsequent send times are all computed prior to measurement by 1178 computing the pseudo-random distribution of inter-packet send times, 1179 (truncating the distribution as specified in the Run-time 1180 Parameters), and the Src sends each packet at the computed times. 1182 6.3.3. Traffic Filtering (observation) Details 1184 The measured results based on a filtered version of the packets 1185 observed, and this section provides the filter details (when 1186 present). 1188
. 1190 NA 1192 6.3.4. Sampling Distribution 1194 1197 NA 1199 6.3.5. Run-time Parameters and Data Format 1201 Run-time Parameters are input factors that must be determined, 1202 configured into the measurement system, and reported with the results 1203 for the context to be complete. 1205 1207 o Src, the IP address of a host (32-bit value for IPv4, 128-bit 1208 value for IPv6) 1210 o Dst, the IP address of a host (32-bit value for IPv4, 128-bit 1211 value for IPv6) 1213 o T0, a time (start of measurement interval, 128-bit NTP Date 1214 Format, see section 6 of [RFC5905]). When T0 is "all-zeros", a 1215 start time is unspecified and Tf is to be interpreted as the 1216 Duration of the measurement interval. 1218 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 1219 see section 6 of [RFC5905]), interpreted as the Duration of the 1220 measurement interval. 1222 o 1/lambda, average packet rate (for Poisson Streams). (1/lambda = 1223 0.1 packet per second, if fixed) 1225 o Upper limit on Poisson distribution (values above this limit will 1226 be clipped and set to the limit value). (if fixed, Upper limit = 1227 300 seconds.) 1229 o ID, the 16-bit identifier assigned by the program that generates 1230 the query, and which must vary in successive queries, see 1231 Section 4.1.1 of [RFC1035]. This identifier is copied into the 1232 corresponding reply and can be used by the requester to match-up 1233 replies to outstanding queries. 1235 The format for 1/lambda and Upper limit of Poisson Dist. are the 1236 short format in [RFC5905] (32 bits) and is as follows: the first 16 1237 bits represent the integer number of seconds; the next 16 bits 1238 represent the fractional part of a second. 1240 >>> should Poisson run-time params be fixed instead? probably yes if 1241 modeling a specific version of MBA tests. 1243 6.3.6. Roles 1245 1247 Src - launches each packet and waits for return transmissions from 1248 Dst. 1250 Dst - waits for each packet from Src and sends a return packet to 1251 Src. 1253 6.4. Output 1255 This category specifies all details of the Output of measurements 1256 using the metric. 1258 6.4.1. Type/Value (two diff terms used) 1260 1262 For all output types: 1264 o T0, a time (start of measurement interval, 128-bit NTP Date 1265 Format, see section 6 of [RFC5905]) 1267 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 1268 see section 6 of [RFC5905]) 1270 Raw -- for each packet sent, pairs of values. 1272 >>> and the status of the response, only assigning values to 1273 successful query-response pairs. 1275 Percentile -- for the conditional distribution of all packets with a 1276 valid value of Round-trip delay (undefined delays are excluded), a 1277 single value corresponding to the 95th percentile. 1279 6.4.2. Data Format 1281 1283 Raw -- for each packet sent, pairs of values as follows: 1285 o T, the time when the packet was sent from Src, 128-bit NTP Date 1286 Format, see section 6 of [RFC5905]) 1288 o dT, a value of Round-trip delay, format is *similar to* the 32-bit 1289 short NTP Time format in Section 6 of [RFC5905] and is as follows: 1290 the first 16 bits represent the *signed* integer number of 1291 seconds; the next 16 bits represent the fractional part of a 1292 second. 1294 o dT is undefined when the packet is not received at Src in waiting 1295 time Tmxax seconds (need undefined code for no-response or un- 1296 successful response) 1298 Percentile -- for the conditional distribution of all packets with a 1299 valid value of Round-trip delay (undefined delays are excluded), a 1300 single value as follows: 1302 See section 4.1 of [RFC3393] for details on the conditional 1303 distribution to exclude undefined values of delay, and Section 5 of 1304 [RFC6703] for background on this analysis choice. 1306 See section 4.3 of [RFC3393] for details on the percentile statistic 1307 (where Round-trip delay should be substituted for "ipdv"). 1309 The percentile = 95. 1311 Data format is a 32-bit signed floating point value, *similar to* the 1312 32-bit short NTP Time format in Section 6 of [RFC5905] and is as 1313 follows: the first 16 bits represent the *signed* integer number of 1314 seconds; the next 16 bits represent the fractional part of a second. 1316 6.4.3. Reference 1318 1320 See the Data Format column for references. 1322 6.4.4. Metric Units 1324 . 1327 Round-trip Delay, dT, is expressed in seconds. 1329 The 95th Percentile of Round-trip Delay is expressed in seconds. 1331 6.5. Administrative items 1333 6.5.1. Status 1335 1337 6.5.2. Requestor (keep?) 1339 name or RFC, etc. 1341 6.5.3. Revision 1343 1.0 1345 6.5.4. Revision Date 1347 YYYY-MM-DD 1349 6.6. Comments and Remarks 1351 Additional (Informational) details for this entry 1353 7. UDP Poisson One-way Delay Registry Entries 1355 This section gives an initial registry entry for the UDP Poisson One- 1356 way Delay. 1358 Note: Each Registry "Name" below specifies a single registry entry, 1359 whose output format varies according to a component of the name that 1360 specifies one form of statistical summary. 1362 IANA is asked to assign a different numeric identifiers to each Name. 1363 All other column entries are the same, thus this section is proposes 1364 five closely-related registry entries. As a result, IANA is also 1365 asked to assign corresponding URIs and URLs. 1367 7.1. Summary 1369 This category includes multiple indexes to the registry entries, the 1370 element ID and metric name. 1372 7.1.1. ID (Identifier) 1374 1377 7.1.2. Name 1379 1381 Act_IP_UDP_Poisson_UDP-Payload-250_One-way_Delay_ 1383 Act_IP_UDP_Poisson_UDP-Payload-250_One-way_Delay_Percentile95 1384 Act_IP_UDP_Poisson_UDP-Payload-250_One-way_Delay_Mean 1386 Act_IP_UDP_Poisson_UDP-Payload-250_One-way_Delay_Min 1388 Act_IP_UDP_Poisson_UDP-Payload-250_One-way_Delay_Max 1390 Act_IP_UDP_Poisson_UDP-Payload-250_One-way_Delay_Std_Dev 1392 7.1.3. URI and URL 1394 URI: Prefix urn:ietf:params:performance:metric... 1396 URL: http:\\www.iana.org\ ... 1398 7.1.4. Description 1400 This metric assesses the delay of a stream of packets exchanged 1401 between two hosts (or measurement points), and reports the 1402 One-way delay for all successfully exchanged packets 1403 based on their conditional delay distribution. 1405 7.2. Metric Definition 1407 This category includes columns to prompt the entry of all necessary 1408 details related to the metric definition, including the RFC reference 1409 and values of input factors, called fixed parameters. 1411 7.2.1. Reference Definition 1413 1415 Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way Delay Metric 1416 for IPPM", RFC 2679, September 1999. 1418 [RFC2679] 1420 Morton, A., and Stephan, E., "Spatial Composition of Metrics", RFC 1421 6049, January 2011. 1423 [RFC6049] 1425 1427 Section 3.4 of [RFC2679] provides the reference definition of the 1428 singleton (single value) One-way delay metric. Section 4.4 of 1429 [RFC2679] provides the reference definition expanded to cover a 1430 multi-value sample. Note that terms such as singleton and sample are 1431 defined in Section 11 of [RFC2330]. 1433 Only successful packet transfers with finite delay are included in 1434 the sample, as prescribed in section 4.1.2 of [RFC6049]. 1436 NOTE: RFC2679 will be replaced by 2679-bis on approval, see draft- 1437 ietf-ippm-2679-bis-01. 1439 7.2.2. Fixed Parameters 1441 1445 Type-P: 1447 o IPv4 header values: 1449 * DSCP: set to 0 1451 * TTL set to 255 1453 * Protocol: Set to 17 (UDP) 1455 o UDP header values: 1457 * Checksum: the checksum must be calculated 1459 o UDP Payload: TWAMP Test Packet Formats, Section 4.1.2 of [RFC5357] 1461 * Security features in use influence the number of Padding 1462 octets. 1464 * 250 octets total, including the TWAMP format 1466 Timeout, Tmax: 3 seconds 1468 7.3. Method of Measurement 1470 This category includes columns for references to relevant sections of 1471 the RFC(s) and any supplemental information needed to ensure an 1472 unambiguous methods for implementations. 1474 7.3.1. Reference Method 1476 1479 The methodology for this metric is defined as Type-P-One-way-Delay- 1480 Poisson-Stream in section 3.6 of [RFC2679] and section 4.6 of 1482 [RFC2679] using the Type-P and Timeout defined under Fixed 1483 Parameters. 1485 The method requires sequence numbers or other send-order information 1486 to be retained at the Src or included with each packet to dis- 1487 ambiguate packet reordering if it occurs. Sequence number is part of 1488 the TWAMP payload described under Fixed Parameters. 1490 7.3.2. Packet Generation Stream 1492 This section gives the details of the packet traffic which is the 1493 basis for measurement. In IPPM metrics, this is called the Stream, 1494 and can easily be dscribed by providing the list of stream 1495 parameters. 1497 1499 Section 11.1.3 of RFC 2681 [RFC2330] provides three methods to 1500 generate Poisson sampling intervals. The reciprocal of lambda is the 1501 average packet rate, thus the Run-time Parameter is 1/lambda. 1503 Method 3 or equivalent SHALL used, where given a start time (Run-time 1504 Parameter), the subsequent send times are all computed prior to 1505 measurement by computing the pseudo-random distribution of inter- 1506 packet send times, (truncating the distribution as specified in the 1507 Run-time Parameters), and the Src sends each packet at the computed 1508 times. 1510 7.3.3. Traffic Filtering (observation) Details 1512 NA 1514 7.3.4. Sampling Distribution 1516 NA 1518 7.3.5. Run-time Parameters and Data Format 1520 Run-time Parameters are input factors that must be determined, 1521 configured into the measurement system, and reported with the results 1522 for the context to be complete. 1524 1526 o Src, the IP address of a host (32-bit value for IPv4, 128-bit 1527 value for IPv6) 1529 o Dst, the IP address of a host (32-bit value for IPv4, 128-bit 1530 value for IPv6) 1532 o T0, a time (start of measurement interval, 128-bit NTP Date 1533 Format, see section 6 of [RFC5905]). When T0 is "all-zeros", a 1534 start time is unspecified and Tf is to be interpreted as the 1535 Duration of the measurement interval. 1537 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 1538 see section 6 of [RFC5905]), interpreted as the Duration of the 1539 measurement interval. 1541 o 1/lambda, average packet rate (for Poisson Streams). (1/lambda = 1542 1 packet per second, if fixed) 1544 o Upper limit on Poisson distribution (values above this limit will 1545 be clipped and set to the limit value). (if fixed, Upper limit = 1546 30 seconds.) 1548 The format for 1/lambda and Upper limit of Poisson Dist. are the 1549 short format in [RFC5905] (32 bits) and is as follows: the first 16 1550 bits represent the integer number of seconds; the next 16 bits 1551 represent the fractional part of a second. 1553 >>> should Poisson run-time params be fixed instead? probably yes if 1554 modeling a specific version of tests. Note in the NAME, i.e. 1555 Poisson3.3 1557 7.3.6. Roles 1559 1561 Src - launches each packet and waits for return transmissions from 1562 Dst. This is the TWAMP Session-Sender. 1564 Dst - waits for each packet from Src and sends a return packet to 1565 Src. This is the TWAMP Session-Reflector. 1567 7.4. Output 1569 This category specifies all details of the Output of measurements 1570 using the metric. 1572 7.4.1. Type/Value (two diff terms used) 1574 1576 See subsection titles below for Types. 1578 7.4.2. Data Format 1580 1582 For all output types --- 1584 o T0, a time (start of measurement interval, 128-bit NTP Date 1585 Format, see section 6 of [RFC5905]) 1587 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 1588 see section 6 of [RFC5905]) 1590 7.4.2.1. Percentile95 1592 The 95th percentile SHALL be calculated using the conditional 1593 distribution of all packets with a finite value of One-way delay 1594 (undefined delays are excluded), a single value as follows: 1596 See section 4.1 of [RFC3393] for details on the conditional 1597 distribution to exclude undefined values of delay, and Section 5 of 1598 [RFC6703] for background on this analysis choice. 1600 See section 4.3 of [RFC3393] for details on the percentile statistic 1601 (where Round-trip delay should be substituted for "ipdv"). 1603 The percentile = 95. 1605 Data format is a 32-bit signed value, *similar to* the 32-bit short 1606 NTP Time format in Section 6 of [RFC5905] and is as follows: the 1607 first 16 bits represent the *signed* integer number of seconds; the 1608 next 16 bits represent the fractional part of a second. 1610 7.4.2.2. Mean 1612 The mean SHALL be calculated using the conditional distribution of 1613 all packets with a finite value of One-way delay (undefined delays 1614 are excluded), a single value as follows: 1616 See section 4.1 of [RFC3393] for details on the conditional 1617 distribution to exclude undefined values of delay, and Section 5 of 1618 [RFC6703] for background on this analysis choice. 1620 See section 4.2.2 of [RFC6049] for details on calculating this 1621 statistic, and 4.2.3 of [RFC6049]. 1623 Data format is a 32-bit signed value, *similar to* the 32-bit short 1624 NTP Time format in Section 6 of [RFC5905] and is as follows: the 1625 first 16 bits represent the *signed* integer number of seconds; the 1626 next 16 bits represent the fractional part of a second. 1628 7.4.2.3. Min 1630 The minimum SHALL be calculated using the conditional distribution of 1631 all packets with a finite value of One-way delay (undefined delays 1632 are excluded), a single value as follows: 1634 See section 4.1 of [RFC3393] for details on the conditional 1635 distribution to exclude undefined values of delay, and Section 5 of 1636 [RFC6703] for background on this analysis choice. 1638 See section 4.3.2 of [RFC6049] for details on calculating this 1639 statistic, and 4.3.3 of [RFC6049]. 1641 Data format is a 32-bit signed value, *similar to* the 32-bit short 1642 NTP Time format in Section 6 of [RFC5905] and is as follows: the 1643 first 16 bits represent the *signed* integer number of seconds; the 1644 next 16 bits represent the fractional part of a second. 1646 7.4.2.4. Max 1648 The maximum SHALL be calculated using the conditional distribution of 1649 all packets with a finite value of One-way delay (undefined delays 1650 are excluded), a single value as follows: 1652 See section 4.1 of [RFC3393] for details on the conditional 1653 distribution to exclude undefined values of delay, and Section 5 of 1654 [RFC6703] for background on this analysis choice. 1656 See section 4.3.2 of [RFC6049] for a closely related method for 1657 calculating this statistic, and 4.3.3 of [RFC6049]. The formula is 1658 as follows: 1660 Max = (FiniteDelay [j]) 1662 such that for some index, j, where 1 <= j <= N 1663 FiniteDelay[j] >= FiniteDelay[n] for all n 1665 Data format is a 32-bit signed value, *similar to* the 32-bit short 1666 NTP Time format in Section 6 of [RFC5905] and is as follows: the 1667 first 16 bits represent the *signed* integer number of seconds; the 1668 next 16 bits represent the fractional part of a second. 1670 7.4.2.5. Std_Dev 1672 7.4.3. Reference 1674 1676 See the Data Format column for references. 1678 7.4.4. Metric Units 1680 . 1683 The of One-way Delay is expressed in seconds. 1685 The 95th Percentile of One-way Delay is expressed in seconds. 1687 7.5. Administrative items 1689 7.5.1. Status 1691 1693 7.5.2. Requestor (keep?) 1695 name or RFC, etc. 1697 7.5.3. Revision 1699 1.0 1701 7.5.4. Revision Date 1703 YYYY-MM-DD 1705 7.6. Comments and Remarks 1707 Additional (Informational) details for this entry 1709 8. UDP Periodic One-way Delay Registry Entries 1711 This section gives an initial registry entry for the UDP Periodic 1712 One-way Delay. 1714 Note: Each Registry "Name" below specifies a single registry entry, 1715 whose output format varies according to a component of the name that 1716 specifies one form of statistical summary. 1718 IANA is asked to assign a different numeric identifiers to each Name. 1719 All other column entries are the same, thus this section is proposes 1720 five closely-related registry entries. As a result, IANA is also 1721 asked to assign corresponding URIs and URLs. 1723 8.1. Summary 1725 This category includes multiple indexes to the registry entries, the 1726 element ID and metric name. 1728 8.1.1. ID (Identifier) 1730 1733 8.1.2. Name 1735 1737 Act_IP_UDP_Periodic-var_UDP-Payload-142_One-way_Delay_ 1739 Act_IP_UDP_Periodic-var_UDP-Payload-142_One-way_Delay_Percentile95 1741 Act_IP_UDP_Periodic-var_UDP-Payload-142_One-way_Delay_Mean 1743 Act_IP_UDP_Periodic-var_UDP-Payload-142_One-way_Delay_Min 1745 Act_IP_UDP_Periodic-var_UDP-Payload-142_One-way_Delay_Max 1747 Act_IP_UDP_Periodic-var_UDP-Payload-142_One-way_Delay_Std_Dev 1749 8.1.3. URI and URL 1751 URI: Prefix urn:ietf:params:performance:metric... 1753 URL: http:\\www.iana.org\ ... 1755 8.1.4. Description 1757 This metric assesses the delay of a stream of packets exchanged 1758 between two hosts (or measurement points), and reports the 1759 One-way delay for all successfully exchanged packets 1760 based on their conditional delay distribution. 1762 8.2. Metric Definition 1764 This category includes columns to prompt the entry of all necessary 1765 details related to the metric definition, including the RFC reference 1766 and values of input factors, called fixed parameters. 1768 8.2.1. Reference Definition 1770 1772 Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way Delay Metric 1773 for IPPM", RFC 2679, September 1999. 1775 [RFC2679] 1777 Morton, A., and Stephan, E., "Spatial Composition of Metrics", RFC 1778 6049, January 2011. 1780 [RFC6049] 1782 1784 Section 3.4 of [RFC2679] provides the reference definition of the 1785 singleton (single value) One-way delay metric. Section 4.4 of 1786 [RFC2679] provides the reference definition expanded to cover a 1787 multi-value sample. Note that terms such as singleton and sample are 1788 defined in Section 11 of [RFC2330]. 1790 Only successful packet transfers with finite delay are included in 1791 the sample, as prescribed in section 4.1.2 of [RFC6049]. 1793 NOTE: RFC2679 will be replaced by 2679-bis on approval, see draft- 1794 ietf-ippm-2679-bis-01. 1796 ANY other conditions, ... 1798 8.2.2. Fixed Parameters 1800 1804 Type-P: 1806 o IPv4 header values: 1808 * DSCP: set to 0 1809 * TTL set to 255 1811 * Protocol: Set to 17 (UDP) 1813 o UDP header values: 1815 * Checksum: the checksum must be calculated 1817 o UDP Payload: TWAMP Test Packet Formats, Section 4.1.2 of [RFC5357] 1819 * Security features in use influence the number of Padding 1820 octets. 1822 * 142 octets total, including the TWAMP format 1824 Timeout, Tmax: 3 seconds 1826 8.3. Method of Measurement 1828 This category includes columns for references to relevant sections of 1829 the RFC(s) and any supplemental information needed to ensure an 1830 unambiguous methods for implementations. 1832 8.3.1. Reference Method 1834 1837 The methodology for this metric is defined as Type-P-One-way-Delay- 1838 Poisson-Stream in section 3.6 of [RFC2679] and section 4.6 of 1839 [RFC2679] using the Type-P and Timeout defined under Fixed 1840 Parameters. 1842 The method requires sequence numbers or other send-order information 1843 to be retained at the Src or included with each packet to dis- 1844 ambiguate packet reordering if it occurs. Sequence number is part of 1845 the TWAMP payload described under Fixed Parameters. 1847 8.3.2. Packet Generation Stream 1849 This section gives the details of the packet traffic which is the 1850 basis for measurement. In IPPM metrics, this is called the Stream, 1851 and can easily be dscribed by providing the list of stream 1852 parameters. 1854 1855 Section 3 of [RFC3432] prescribes the method for generating Periodic 1856 streams using associated parameters. 1858 o incT, the nominal duration of inter-packet interval, first bit to 1859 first bit 1861 o dT, the duration of the interval for allowed sample start times 1863 o T0, the actual start time 1865 NOTE: an initiation process with a number of control exchanges 1866 resulting in unpredictable start times (within a time interval) may 1867 be sufficient to avoid synchronization of periodic streams, and 1868 therefore a valid replacement for selecting a start time at random 1869 from a fixed interval. 1871 These stream parameters will be specified as Run-time parameters. 1873 8.3.3. Traffic Filtering (observation) Details 1875 NA 1877 8.3.4. Sampling Distribution 1879 NA 1881 8.3.5. Run-time Parameters and Data Format 1883 Run-time Parameters are input factors that must be determined, 1884 configured into the measurement system, and reported with the results 1885 for the context to be complete. 1887 1889 o Src, the IP address of a host (32-bit value for IPv4, 128-bit 1890 value for IPv6) 1892 o Dst, the IP address of a host (32-bit value for IPv4, 128-bit 1893 value for IPv6) 1895 o T0, a time (start of measurement interval, 128-bit NTP Date 1896 Format, see section 6 of [RFC5905]). When T0 is "all-zeros", a 1897 start time is unspecified and Tf is to be interpreted as the 1898 Duration of the measurement interval. 1900 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 1901 see section 6 of [RFC5905]), interpreted as the Duration of the 1902 measurement interval. 1904 o incT, the nominal duration of inter-packet interval, first bit to 1905 first bit 1907 o dT, the duration of the interval for allowed sample start times 1909 The format for incT and dT are the short format in [RFC5905] (32 1910 bits) and is as follows: the first 16 bits represent the integer 1911 number of seconds; the next 16 bits represent the fractional part of 1912 a second. 1914 >>> should Periodic run-time params be fixed instead? probably yes if 1915 modeling a specific version of tests. Note in the NAME, i.e. 1916 Poisson3.3 1918 8.3.6. Roles 1920 1922 Src - launches each packet and waits for return transmissions from 1923 Dst. This is the TWAMP Session-Sender. 1925 Dst - waits for each packet from Src and sends a return packet to 1926 Src. This is the TWAMP Session-Reflector. 1928 8.4. Output 1930 This category specifies all details of the Output of measurements 1931 using the metric. 1933 8.4.1. Type/Value (two diff terms used) 1935 1937 See subsection titles in Data Format for Types. 1939 8.4.2. Data Format 1941 1943 For all output types --- 1945 o T0, a time (start of measurement interval, 128-bit NTP Date 1946 Format, see section 6 of [RFC5905]) 1948 o Tf, a time (end of measurement interval, 128-bit NTP Date Format, 1949 see section 6 of [RFC5905]) 1951 8.4.2.1. Percentile95 1953 The 95th percentile SHALL be calculated using the conditional 1954 distribution of all packets with a finite value of One-way delay 1955 (undefined delays are excluded), a single value as follows: 1957 See section 4.1 of [RFC3393] for details on the conditional 1958 distribution to exclude undefined values of delay, and Section 5 of 1959 [RFC6703] for background on this analysis choice. 1961 See section 4.3 of [RFC3393] for details on the percentile statistic 1962 (where Round-trip delay should be substituted for "ipdv"). 1964 The percentile = 95. 1966 Data format is a 32-bit signed value, *similar to* the 32-bit short 1967 NTP Time format in Section 6 of [RFC5905] and is as follows: the 1968 first 16 bits represent the *signed* integer number of seconds; the 1969 next 16 bits represent the fractional part of a second. 1971 8.4.2.2. Mean 1973 The mean SHALL be calculated using the conditional distribution of 1974 all packets with a finite value of One-way delay (undefined delays 1975 are excluded), a single value as follows: 1977 See section 4.1 of [RFC3393] for details on the conditional 1978 distribution to exclude undefined values of delay, and Section 5 of 1979 [RFC6703] for background on this analysis choice. 1981 See section 4.2.2 of [RFC6049] for details on calculating this 1982 statistic, and 4.2.3 of [RFC6049]. 1984 Data format is a 32-bit signed value, *similar to* the 32-bit short 1985 NTP Time format in Section 6 of [RFC5905] and is as follows: the 1986 first 16 bits represent the *signed* integer number of seconds; the 1987 next 16 bits represent the fractional part of a second. 1989 8.4.2.3. Min 1991 The minimum SHALL be calculated using the conditional distribution of 1992 all packets with a finite value of One-way delay (undefined delays 1993 are excluded), a single value as follows: 1995 See section 4.1 of [RFC3393] for details on the conditional 1996 distribution to exclude undefined values of delay, and Section 5 of 1997 [RFC6703] for background on this analysis choice. 1999 See section 4.3.2 of [RFC6049] for details on calculating this 2000 statistic, and 4.3.3 of [RFC6049]. 2002 Data format is a 32-bit signed value, *similar to* the 32-bit short 2003 NTP Time format in Section 6 of [RFC5905] and is as follows: the 2004 first 16 bits represent the *signed* integer number of seconds; the 2005 next 16 bits represent the fractional part of a second. 2007 8.4.2.4. Max 2009 The maximum SHALL be calculated using the conditional distribution of 2010 all packets with a finite value of One-way delay (undefined delays 2011 are excluded), a single value as follows: 2013 See section 4.1 of [RFC3393] for details on the conditional 2014 distribution to exclude undefined values of delay, and Section 5 of 2015 [RFC6703] for background on this analysis choice. 2017 See section 4.3.2 of [RFC6049] for a closely related method for 2018 calculating this statistic, and 4.3.3 of [RFC6049]. The formula is 2019 as follows: 2021 Max = (FiniteDelay [j]) 2023 such that for some index, j, where 1 <= j <= N 2024 FiniteDelay[j] >= FiniteDelay[n] for all n 2026 Data format is a 32-bit signed value, *similar to* the 32-bit short 2027 NTP Time format in Section 6 of [RFC5905] and is as follows: the 2028 first 16 bits represent the *signed* integer number of seconds; the 2029 next 16 bits represent the fractional part of a second. 2031 8.4.2.5. Std_Dev 2033 8.4.3. Reference 2035 2037 See the Data Format column for references. 2039 8.4.4. Metric Units 2041 . 2044 The of One-way Delay is expressed in seconds. 2046 8.5. Administrative items 2048 8.5.1. Status 2050 2052 8.5.2. Requestor (keep?) 2054 name or RFC, etc. 2056 8.5.3. Revision 2058 1.0 2060 8.5.4. Revision Date 2062 YYYY-MM-DD 2064 8.6. Comments and Remarks 2066 Additional (Informational) details for this entry 2068 9. partly BLANK Registry Entry 2070 This section gives an initial registry entry for .... 2072 9.1. Summary 2074 This category includes multiple indexes to the registry entries, the 2075 element ID and metric name. 2077 2079 9.1.1. ID (Identifier) 2081 2083 9.1.2. Name 2085 2087 URL: ?? 2089 9.1.3. URI 2091 URI: Prefix urn:ietf:params:performance:metric 2093 9.1.4. Description 2095 TBD. 2097 9.2. Metric Definition 2099 This category includes columns to prompt the entry of all necessary 2100 details related to the metric definition, including the RFC reference 2101 and values of input factors, called fixed parameters. 2103 9.2.1. Reference Definition 2105 2107 Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip Delay 2108 Metric for IPPM", RFC 2681, September 1999. 2110 2112 Section 2.4 of [RFC2681] provides the reference definition of the 2113 singleton (single value) Round-trip delay metric. Section 3.4 of 2114 [RFC2681] provides the reference definition expanded to cover a 2115 multi-value sample. Note that terms such as singleton and sample are 2116 defined in Section 11 of [RFC2330]. 2118 Note that although the definition of "Round-trip-Delay between Src 2119 and Dst at T" is directionally ambiguous in the text, this metric 2120 tightens the definition further to recognize that the host in the 2121 "Src" role will send the first packet to "Dst", and ultimately 2122 receive the corresponding return packet from "Dst" (when neither are 2123 lost). 2125 <<< Check how the Methodology also makes this clear (or not) >>> 2127 9.2.2. Fixed Parameters 2129 2133 Type-P: 2135 o IPv4 header values: 2137 * DSCP: set to 0 2139 * TTL set to 255 2140 * Protocol: Set to 17 (UDP) 2142 o UDP header values: 2144 * Checksum: the checksum must be calculated 2146 o Payload 2148 * Sequence number: 8-byte integer 2150 * Timestamp: 8 byte integer. Expressed as 64-bit NTP timestamp 2151 as per section 6 of RFC 5905 [RFC5905] 2153 * No padding (total of 9 bytes) 2155 Timeout: 3 seconds 2157 9.3. Method of Measurement 2159 This category includes columns for references to relevant sections of 2160 the RFC(s) and any supplemental information needed to ensure an 2161 unambiguous methods for implementations. 2163 9.3.1. Reference Method 2165 2168 9.3.2. Packet Generation Stream 2170 This section gives the details of the packet traffic which is the 2171 basis for measurement. In IPPM metrics, this is called the Stream, 2172 and can easily be dscribed by providing the list of stream 2173 parameters. 2175 2177 9.3.3. Traffic Filtering (observation) Details 2179 The measured results based on a filtered version of the packets 2180 observed, and this section provides the filter details (when 2181 present). 2183
. 2185 9.3.4. Sampling Distribution 2187 2190 9.3.5. Run-time Parameters and Data Format 2192 Run-time Parameters are input factors that must be determined, 2193 configured into the measurement system, and reported with the results 2194 for the context to be complete. 2196 2198 . 2200 9.3.6. Roles 2202 2204 9.4. Output 2206 This category specifies all details of the Output of measurements 2207 using the metric. 2209 9.4.1. Type/Value (two diff terms used) 2211 2213 9.4.2. Data Format 2215 2217 o Value: 2219 o Data Format: (There may be some precedent to follow here, but 2220 otherwise use 64-bit NTP Timestamp Format, see section 6 of 2221 [RFC5905]). 2223 o Reference:
2225 9.4.3. Reference 2227 2229 9.4.4. Metric Units 2231 . 2234 9.5. Administrative items 2236 9.5.1. Status 2238 2240 9.5.2. Requestor (keep?) 2242 name or RFC, etc. 2244 9.5.3. Revision 2246 1.0 2248 9.5.4. Revision Date 2250 YYYY-MM-DD 2252 9.6. Comments and Remarks 2254 Additional (Informational) details for this entry 2256 10. BLANK Registry Entry 2258 This section gives an initial registry entry for .... 2260 10.1. Summary 2262 This category includes multiple indexes to the registry entries, the 2263 element ID and metric name. 2265 2267 10.1.1. ID (Identifier) 2269 2271 10.1.2. Name 2273 2275 URL: ?? 2277 10.1.3. URI 2279 URI: Prefix urn:ietf:params:performance:metric 2281 10.1.4. Description 2283 TBD. 2285 10.2. Metric Definition 2287 This category includes columns to prompt the entry of all necessary 2288 details related to the metric definition, including the RFC reference 2289 and values of input factors, called fixed parameters. 2291 10.2.1. Reference Definition 2293 2295 2297 10.2.2. Fixed Parameters 2299 2303 10.3. Method of Measurement 2305 This category includes columns for references to relevant sections of 2306 the RFC(s) and any supplemental information needed to ensure an 2307 unambiguous methods for implementations. 2309 10.3.1. Reference Method 2311 2314 10.3.2. Packet Generation Stream 2316 2318 10.3.3. Traffic Filtering (observation) Details 2320 . 2324 10.3.4. Sampling Distribution 2326 2329 10.3.5. Run-time Parameters and Data Format 2331 . 2333 10.3.6. Roles 2335 2337 10.4. Output 2339 This category specifies all details of the Output of measurements 2340 using the metric. 2342 10.4.1. Type/Value (two diff terms used) 2344 2346 10.4.2. Data Format 2348 2350 10.4.3. Reference 2352 2354 10.4.4. Metric Units 2356 . 2359 10.5. Administrative items 2361 10.5.1. Status 2363 2365 10.5.2. Requestor (keep?) 2367 2369 10.5.3. Revision 2371 1.0 2373 10.5.4. Revision Date 2375 YYYY-MM-DD 2377 10.6. Comments and Remarks 2379 Additional (Informational) details for this entry 2381 11. Example RTCP-XR Registry Entry 2383 This section is MAY BE DELETED or adapted before submission. 2385 This section gives an example registry entry for the end-point metric 2386 described in RFC 7003 [RFC7003], for RTCP-XR Burst/Gap Discard Metric 2387 reporting. 2389 11.1. Registry Indexes 2391 This category includes multiple indexes to the registry entries, the 2392 element ID and metric name. 2394 11.1.1. Identifier 2396 An integer having enough digits to uniquely identify each entry in 2397 the Registry. 2399 11.1.2. Name 2401 A metric naming convention is TBD. 2403 11.1.3. URI 2405 Prefix urn:ietf:params:performance:metric 2407 11.1.4. Status 2409 current 2411 11.1.5. Requestor 2413 Alcelip Mornuley 2415 11.1.6. Revision 2417 1.0 2419 11.1.7. Revision Date 2421 2014-07-04 2423 11.1.8. Description 2425 TBD. 2427 11.1.9. Reference Specification(s) 2429 [RFC3611][RFC4566][RFC6776][RFC6792][RFC7003] 2431 11.2. Metric Definition 2433 This category includes columns to prompt the entry of all necessary 2434 details related to the metric definition, including the RFC reference 2435 and values of input factors, called fixed parameters. Section 3.2 of 2436 [RFC7003] provides the reference information for this category. 2438 11.2.1. Reference Definition 2440 Packets Discarded in Bursts: 2442 The total number of packets discarded during discard bursts. The 2443 measured value is unsigned value. If the measured value exceeds 2444 0xFFFFFD, the value 0xFFFFFE MUST be reported to indicate an over- 2445 range measurement. If the measurement is unavailable, the value 2446 0xFFFFFF MUST be reported. 2448 11.2.2. Fixed Parameters 2450 Fixed Parameters are input factors that must be determined and 2451 embedded in the measurement system for use when needed. The values 2452 of these parameters is specified in the Registry. 2454 Threshold: 8 bits, set to value = 3 packets. 2456 The Threshold is equivalent to Gmin in [RFC3611], i.e., the number of 2457 successive packets that must not be discarded prior to and following 2458 a discard packet in order for this discarded packet to be regarded as 2459 part of a gap. Note that the Threshold is set in accordance with the 2460 Gmin calculation defined in Section 4.7.2 of [RFC3611]. 2462 Interval Metric flag: 2 bits, set to value 11=Cumulative Duration 2463 This field is used to indicate whether the burst/gap discard metrics 2464 are Sampled, Interval, or Cumulative metrics [RFC6792]: 2466 I=10: Interval Duration - the reported value applies to the most 2467 recent measurement interval duration between successive metrics 2468 reports. 2470 I=11: Cumulative Duration - the reported value applies to the 2471 accumulation period characteristic of cumulative measurements. 2473 Senders MUST NOT use the values I=00 or I=01. 2475 11.3. Method of Measurement 2477 This category includes columns for references to relevant sections of 2478 the RFC(s) and any supplemental information needed to ensure an 2479 unambiguous methods for implementations. For the Burst/Gap Discard 2480 Metric, it appears that the only guidance on methods of measurement 2481 is in Section 3.0 of [RFC7003] and its supporting references. 2482 Relevant information is repeated below, although there appears to be 2483 no section titled "Method of Measurement" in [RFC7003]. 2485 11.3.1. Reference Method 2487 Metrics in this block report on burst/gap discard in the stream 2488 arriving at the RTP system. Measurements of these metrics are made 2489 at the receiving end of the RTP stream. Instances of this metrics 2490 block use the synchronization source (SSRC) to refer to the separate 2491 auxiliary Measurement Information Block [RFC6776], which describes 2492 measurement periods in use (see [RFC6776], Section 4.2). 2494 This metrics block relies on the measurement period in the 2495 Measurement Information Block indicating the span of the report. 2496 Senders MUST send this block in the same compound RTCP packet as the 2497 Measurement Information Block. Receivers MUST verify that the 2498 measurement period is received in the same compound RTCP packet as 2499 this metrics block. If not, this metrics block MUST be discarded. 2501 11.3.2. Stream Type and Stream Parameters 2503 Since RTCP-XR Measurements are conducted on live RTP traffic, the 2504 complete description of the stream is contained in SDP messages that 2505 proceed the establishment of a compatible stream between two or more 2506 communicating hosts. See Run-time Parameters, below. 2508 11.3.3. Output Type and Data Format 2510 The output type defines the type of result that the metric produces. 2512 o Value: Packets Discarded in Bursts 2514 o Data Format: 24 bits 2516 o Reference: Section 3.2 of [RFC7003] 2518 11.3.4. Metric Units 2520 The measured results are apparently expressed in packets, although 2521 there is no section of [RFC7003] titled "Metric Units". 2523 11.3.5. Run-time Parameters and Data Format 2525 Run-Time Parameters are input factors that must be determined, 2526 configured into the measurement system, and reported with the results 2527 for the context to be complete. However, the values of these 2528 parameters is not specified in the Registry, rather these parameters 2529 are listed as an aid to the measurement system implementor or user 2530 (they must be left as variables, and supplied on execution). 2532 The Data Format of each Run-time Parameter SHALL be specified in this 2533 column, to simplify the control and implementation of measurement 2534 devices. 2536 SSRC of Source: 32 bits As defined in Section 4.1 of [RFC3611]. 2538 SDP Parameters: As defined in [RFC4566] 2540 Session description v= (protocol version number, currently only 0) 2542 o= (originator and session identifier : username, id, version number, 2543 network address) 2545 s= (session name : mandatory with at least one UTF-8-encoded 2546 character) 2548 i=* (session title or short information) u=* (URI of description) 2550 e=* (zero or more email address with optional name of contacts) 2552 p=* (zero or more phone number with optional name of contacts) 2554 c=* (connection information--not required if included in all media) 2555 b=* (zero or more bandwidth information lines) One or more Time 2556 descriptions ("t=" and "r=" lines; see below) 2558 z=* (time zone adjustments) 2560 k=* (encryption key) 2562 a=* (zero or more session attribute lines) 2564 Zero or more Media descriptions (each one starting by an "m=" line; 2565 see below) 2567 m= (media name and transport address) 2569 i=* (media title or information field) 2571 c=* (connection information -- optional if included at session level) 2573 b=* (zero or more bandwidth information lines) 2575 k=* (encryption key) 2577 a=* (zero or more media attribute lines -- overriding the Session 2578 attribute lines) 2580 An example Run-time SDP description follows: 2582 v=0 2584 o=jdoe 2890844526 2890842807 IN IP4 192.0.2.5 2586 s=SDP Seminar i=A Seminar on the session description protocol 2588 u=http://www.example.com/seminars/sdp.pdf e=j.doe@example.com (Jane 2589 Doe) 2591 c=IN IP4 233.252.0.12/127 2593 t=2873397496 2873404696 2595 a=recvonly 2597 m=audio 49170 RTP/AVP 0 2599 m=video 51372 RTP/AVP 99 2601 a=rtpmap:99 h263-1998/90000 2603 11.4. Comments and Remarks 2605 TBD. 2607 12. Security Considerations 2609 These registry entries represent no known security implications for 2610 Internet Security. Each referenced Metric contains a Security 2611 Considerations section. 2613 13. IANA Considerations 2615 IANA is requested to populate The Performance Metric Registry defined 2616 in [I-D.ietf-ippm-metric-registry] with the values defined above. 2618 2620 14. Acknowledgements 2622 The authors thank Brian Trammell for suggesting the term "Run-time 2623 Parameters", which led to the distinction between run-time and fixed 2624 parameters implemented in this memo, for identifying the IPFIX metric 2625 with Flow Key as an example, and for many other productive 2626 suggestions. Thanks to Peter Koch, who provided several useful 2627 suggestions for disambiguating successive DNS Queries in the DNS 2628 Response time metric. 2630 The authors also acknowledge the constructive reviews and helpful 2631 suggestions from Barbara Stark, Juergen Schoenwaelder, Tim Carey, and 2632 participants in the LMAP working group. 2634 15. References 2636 15.1. Normative References 2638 [I-D.ietf-ippm-metric-registry] 2639 Bagnulo, M., Claise, B., Eardley, P., and A. Morton, 2640 "Registry for Performance Metrics", Internet Draft (work 2641 in progress) draft-ietf-ippm-metric-registry, 2014. 2643 [RFC1035] Mockapetris, P., "Domain names - implementation and 2644 specification", STD 13, RFC 1035, DOI 10.17487/RFC1035, 2645 November 1987, . 2647 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 2648 Requirement Levels", BCP 14, RFC 2119, 2649 DOI 10.17487/RFC2119, March 1997, 2650 . 2652 [RFC2330] Paxson, V., Almes, G., Mahdavi, J., and M. Mathis, 2653 "Framework for IP Performance Metrics", RFC 2330, 2654 DOI 10.17487/RFC2330, May 1998, 2655 . 2657 [RFC2679] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way 2658 Delay Metric for IPPM", RFC 2679, DOI 10.17487/RFC2679, 2659 September 1999, . 2661 [RFC2680] Almes, G., Kalidindi, S., and M. Zekauskas, "A One-way 2662 Packet Loss Metric for IPPM", RFC 2680, 2663 DOI 10.17487/RFC2680, September 1999, 2664 . 2666 [RFC2681] Almes, G., Kalidindi, S., and M. Zekauskas, "A Round-trip 2667 Delay Metric for IPPM", RFC 2681, DOI 10.17487/RFC2681, 2668 September 1999, . 2670 [RFC3393] Demichelis, C. and P. Chimento, "IP Packet Delay Variation 2671 Metric for IP Performance Metrics (IPPM)", RFC 3393, 2672 DOI 10.17487/RFC3393, November 2002, 2673 . 2675 [RFC3432] Raisanen, V., Grotefeld, G., and A. Morton, "Network 2676 performance measurement with periodic streams", RFC 3432, 2677 DOI 10.17487/RFC3432, November 2002, 2678 . 2680 [RFC4737] Morton, A., Ciavattone, L., Ramachandran, G., Shalunov, 2681 S., and J. Perser, "Packet Reordering Metrics", RFC 4737, 2682 DOI 10.17487/RFC4737, November 2006, 2683 . 2685 [RFC5357] Hedayat, K., Krzanowski, R., Morton, A., Yum, K., and J. 2686 Babiarz, "A Two-Way Active Measurement Protocol (TWAMP)", 2687 RFC 5357, DOI 10.17487/RFC5357, October 2008, 2688 . 2690 [RFC5905] Mills, D., Martin, J., Ed., Burbank, J., and W. Kasch, 2691 "Network Time Protocol Version 4: Protocol and Algorithms 2692 Specification", RFC 5905, DOI 10.17487/RFC5905, June 2010, 2693 . 2695 [RFC6049] Morton, A. and E. Stephan, "Spatial Composition of 2696 Metrics", RFC 6049, DOI 10.17487/RFC6049, January 2011, 2697 . 2699 [RFC6673] Morton, A., "Round-Trip Packet Loss Metrics", RFC 6673, 2700 DOI 10.17487/RFC6673, August 2012, 2701 . 2703 15.2. Informative References 2705 [Brow00] Brownlee, N., "Packet Matching for NeTraMet 2706 Distributions", March 2000. 2708 [RFC1242] Bradner, S., "Benchmarking Terminology for Network 2709 Interconnection Devices", RFC 1242, DOI 10.17487/RFC1242, 2710 July 1991, . 2712 [RFC3611] Friedman, T., Ed., Caceres, R., Ed., and A. Clark, Ed., 2713 "RTP Control Protocol Extended Reports (RTCP XR)", 2714 RFC 3611, DOI 10.17487/RFC3611, November 2003, 2715 . 2717 [RFC4148] Stephan, E., "IP Performance Metrics (IPPM) Metrics 2718 Registry", BCP 108, RFC 4148, DOI 10.17487/RFC4148, August 2719 2005, . 2721 [RFC4566] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session 2722 Description Protocol", RFC 4566, DOI 10.17487/RFC4566, 2723 July 2006, . 2725 [RFC5472] Zseby, T., Boschi, E., Brownlee, N., and B. Claise, "IP 2726 Flow Information Export (IPFIX) Applicability", RFC 5472, 2727 DOI 10.17487/RFC5472, March 2009, 2728 . 2730 [RFC5477] Dietz, T., Claise, B., Aitken, P., Dressler, F., and G. 2731 Carle, "Information Model for Packet Sampling Exports", 2732 RFC 5477, DOI 10.17487/RFC5477, March 2009, 2733 . 2735 [RFC5481] Morton, A. and B. Claise, "Packet Delay Variation 2736 Applicability Statement", RFC 5481, DOI 10.17487/RFC5481, 2737 March 2009, . 2739 [RFC6248] Morton, A., "RFC 4148 and the IP Performance Metrics 2740 (IPPM) Registry of Metrics Are Obsolete", RFC 6248, 2741 DOI 10.17487/RFC6248, April 2011, 2742 . 2744 [RFC6390] Clark, A. and B. Claise, "Guidelines for Considering New 2745 Performance Metric Development", BCP 170, RFC 6390, 2746 DOI 10.17487/RFC6390, October 2011, 2747 . 2749 [RFC6703] Morton, A., Ramachandran, G., and G. Maguluri, "Reporting 2750 IP Network Performance Metrics: Different Points of View", 2751 RFC 6703, DOI 10.17487/RFC6703, August 2012, 2752 . 2754 [RFC6776] Clark, A. and Q. Wu, "Measurement Identity and Information 2755 Reporting Using a Source Description (SDES) Item and an 2756 RTCP Extended Report (XR) Block", RFC 6776, 2757 DOI 10.17487/RFC6776, October 2012, 2758 . 2760 [RFC6792] Wu, Q., Ed., Hunt, G., and P. Arden, "Guidelines for Use 2761 of the RTP Monitoring Framework", RFC 6792, 2762 DOI 10.17487/RFC6792, November 2012, 2763 . 2765 [RFC7003] Clark, A., Huang, R., and Q. Wu, Ed., "RTP Control 2766 Protocol (RTCP) Extended Report (XR) Block for Burst/Gap 2767 Discard Metric Reporting", RFC 7003, DOI 10.17487/RFC7003, 2768 September 2013, . 2770 [RFC7594] Eardley, P., Morton, A., Bagnulo, M., Burbridge, T., 2771 Aitken, P., and A. Akhter, "A Framework for Large-Scale 2772 Measurement of Broadband Performance (LMAP)", RFC 7594, 2773 DOI 10.17487/RFC7594, September 2015, 2774 . 2776 Authors' Addresses 2778 Al Morton 2779 AT&T Labs 2780 200 Laurel Avenue South 2781 Middletown,, NJ 07748 2782 USA 2784 Phone: +1 732 420 1571 2785 Fax: +1 732 368 1192 2786 Email: acmorton@att.com 2787 URI: http://home.comcast.net/~acmacm/ 2788 Marcelo Bagnulo 2789 Universidad Carlos III de Madrid 2790 Av. Universidad 30 2791 Leganes, Madrid 28911 2792 SPAIN 2794 Phone: 34 91 6249500 2795 Email: marcelo@it.uc3m.es 2796 URI: http://www.it.uc3m.es 2798 Philip Eardley 2799 BT 2800 Adastral Park, Martlesham Heath 2801 Ipswich 2802 ENGLAND 2804 Email: philip.eardley@bt.com 2806 Kevin D'Souza 2807 AT&T Labs 2808 200 Laurel Avenue South 2809 Middletown,, NJ 07748 2810 USA 2812 Phone: +1 732 420 xxxx 2813 Email: kevin.dsouza@att.com