Transcript bmwg-5
Content-Aware Device Benchmarking Methodology (draft-hamilton-bmwg-ca-bench-meth-05) BMWG Meeting IETF-79 Beijing November 2010 Mike Hamilton [email protected] BreakingPoint Systems 1 Agenda Charter Update Now What? Proposed Schedule Goals Reset Goals/Non-Goals Proposed Schedule Outstanding Questions Draft-05 highlights 2 Charter Update New classes of network devices that operate above the IP layer of the network stack require a new methodology to perform adequate benchmarking. Existing BMWG RFCs (RFC2647 and RFC3511) provides useful measurement and performance statistics, though they may not reflect the actual performance of the device when deployed in production networks. Operating within the limitations of the charter, namely blackbox characterization in laboratory environments, the BMWG will develop a methodology that more closely relates the performance of these devices to performance in an operational setting. In order to confirm or identify key performance characteristics, BMWG will solicit input from operations groups such as NANOG, RIP and APRICOT. 3 Now What? •Define strategy for getting the work done • Terminology Draft • Methodology Draft •Call for volunteers • Would like to setup weekly meetings 4 Proposed Schedule •December 2011 • Terminology Draft for IESG Review • Methodology Draft for IESG Review •December 2010 • -00 terminology draft •January 2011 • -00 methodology draft •March – June 2011 • Solicit WG feedback •June 2011 • Start WGLC 5 Goals Reset • Create a series of benchmark tests to MOST accurately predict device performance under realistic conditions FOR A SPECIFIC SIMULATED NETWORK • RFC 2544 Quotes Page 11, Section 18, “Multiple Frame Sizes” • “The distribution MAY approximate the conditions on the network in which the DUT would be used.” • “The authors do not have any idea how the results of such a test would be interpreted other than to directly compare multiple DUTs in some very specific simulated network” 6 Goals/Non-Goals •Goals • Repeatable Results • Random/Repeatable • Testbed to testbed • Compare Multiple DUTs •Non-Goals • Replace RFC 2544/3511 7 Outstanding Questions • Traffic Composition • Basket? • Methodology for selecting? • Updatable? • One network is different from another every day •Security? • Is it worth the trouble? •Malformed Traffic •What information is necessary to create repeatable ‘application layer’ flows? • Type of flow, size of flow, etc •Others? 8 Draft-05 Highlights and Reasons • “Shell” Methodology • More reproducible •Added back ‘security’ • Why shy away when application layer already difficult •Maintain ‘fuzzing’ aspect • Random but repeatable 9