{"draft":"draft-ietf-bmwg-benchmarking-stateful-09","doc_id":"RFC9693","title":"Benchmarking Methodology for Stateful NATxy Gateways","authors":["G. Lencse","K. Shima"],"format":["HTML","TEXT","PDF","XML"],"page_count":"24","pub_status":"INFORMATIONAL","status":"INFORMATIONAL","source":"Benchmarking Methodology","abstract":"RFC 2544 defines a benchmarking methodology for network interconnect\r\ndevices. RFC 5180 addresses IPv6 specificities, and it also provides\r\na technology update but excludes IPv6 transition technologies. RFC\r\n8219 addresses IPv6 transition technologies, including stateful\r\nNAT64. However, none of them discuss how to apply pseudorandom port\r\nnumbers from RFC 4814 to any stateful NATxy (such as NAT44, NAT64,\r\nand NAT66) technologies.  This document discusses why using\r\npseudorandom port numbers with stateful NATxy gateways is a difficult\r\nproblem. It recommends a solution that limits the port number ranges\r\nand uses two test phases (phase 1 and phase 2). This document shows\r\nhow the classic performance measurement procedures (e.g., throughput,\r\nframe loss rate, latency, etc.)  can be carried out.  New performance\r\nmetrics and measurement procedures are also defined for measuring the\r\nmaximum connection establishment rate, connection tear-down rate, and\r\nconnection tracking table capacity.","pub_date":"January 2025","keywords":["Benchmarking","Stateful NATxy","Measurement Procedure","Throughput","Frame Loss Rate","Latency","PDV"],"obsoletes":[],"obsoleted_by":[],"updates":[],"updated_by":[],"see_also":[],"doi":"10.17487\/RFC9693","errata_url":null}