310 likes | 403 Views
Differentiated Service Test REPORT. TF-TANT Tiziana Ferrari ferrari@cnaf.infn.it Frankfurt, 1 Oct 1999. Overview. Differentiated Services Architecture test activities test infrastructure interim results QoS measurement test plan Services. Differentiated Services Architecture.
E N D
Differentiated Service TestREPORT TF-TANT Tiziana Ferrari ferrari@cnaf.infn.it Frankfurt, 1 Oct 1999 Tiziana Ferrari Differentiated Services Test: Report
Overview • Differentiated Services Architecture • test activities • test infrastructure • interim results • QoS measurement test plan • Services Tiziana Ferrari Differentiated Services Test: Report
Differentiated ServicesArchitecture Tiziana Ferrari Differentiated Services Test: Report
Differentiated services cont. • Why diffserv? • new technology • independent of layer 2 technologies • interoperability between independent national research networks (different requirements, infrastructures, policies and management) • traffic aggregation • scalability • nosignalling • QoS for networks not ATM based Tiziana Ferrari Differentiated Services Test: Report
Diffserv architecture: components • DS field (1 byte) • packet classification • packet scheduling • traffic conditioning: • metering • marking • policing • shaping 0 6 7 CU DS field Tiziana Ferrari Differentiated Services Test: Report
Diffserv architecture: PHBs • Per Hop Behaviours: description of the packet treatment into the network • Standardised PHBs: • Expedited Forwarding (low delay, low delay variation, guaranteed bandwidth) • Assured Forwarding (Behaviour Aggregate, 4 classes, 3 drop priorities per class) • Experimental PHBs • PHB class selectors: XXX000 (backward compatibility with TOS precedence) Tiziana Ferrari Differentiated Services Test: Report
Diffserv architecture: network model DS domain NRN DS domain MPLS Non DS capable domain DS domain NRN Marking policing scheduling TEN-155 marking shaping DS domain DS domain Tiziana Ferrari Differentiated Services Test: Report
Test activities: test plan current status future tests Tiziana Ferrari Differentiated Services Test: Report
Programme overview • Diffserv architecture (precedence, DSCP) • test of QoS features in routers • parameter tuning • interoperability • mixed diffserv-intserv architecture (end-to-end QoS) • mixed diffserv-MPLS architecture • QoS measurement • Policy management Tiziana Ferrari Differentiated Services Test: Report
Work accomplished (from end of June 99) • network configuration • baseline performance testing • CISCO: • CAR (Committed Access Rate): functionality and parameter tuning • CB-WFQ (Class Based WFQ): traffic isolation • IBM: • policing of EF traffic • SCFQ (Self Clocked Fair Queuing): EF, AF and best-effort • configuration of GPS capable QoS measurement equipment (SmartBits by Netcom Systems) • diffserv deployment by a test application: MONARC • definition of test services • definition of QoS measurement test plan Tiziana Ferrari Differentiated Services Test: Report
Future activities • Weighted Random Early Discard (WRED) testing • completion of IBM testing • study of TCP related issues: ACK colouring, TCP and UDP aggregations with AF • interoperability testing (other future platforms: NORTEL, TELEBIT, TORRENT, Linux) • service testing • QoS performance measurement • diffserv and MPLS • diffserv and intserv • policy management and bandwidth brokerage Tiziana Ferrari Differentiated Services Test: Report
Test infrastructure Tiziana Ferrari Differentiated Services Test: Report
Test network Tiziana Ferrari Differentiated Services Test: Report
LAN layout (example) Tiziana Ferrari Differentiated Services Test: Report
Equipment • Test workstations • traffic generators (3 SmartBits, Netcom System loan) • ATM switches in the test sites • cabletron ethernet switch • DS capable routers: • C7200 or C7500 (partial CISCO loan) • IBM 2212 and IBM 2216 (2 routers in 5 sites, IBM donation) Tiziana Ferrari Differentiated Services Test: Report
Interim Results Tiziana Ferrari Differentiated Services Test: Report
CISCO: CAR • Multi-field classification: OK • packet marking (precedence setting): OK • exceed action testing: OK • policing (at a configurable rate): two parameters token bucket for TCP performance optimisation • normal burst • exceed burst --> parameter tuning Tiziana Ferrari Differentiated Services Test: Report
CAR: test of exceed actions Throughput: SWITCH: 1.20 Mbps DANTE: 0.38 Mbps Tiziana Ferrari Differentiated Services Test: Report
CAR: TCP burst tolerance • Normal and exceed burst tuning with TCP traffic • single and multiple TCP streams optimum values are functions of the rate R at which traffic is policed, in particular: normal burst = 0.5 sec * R exceed burst = 2 * normal burst Tiziana Ferrari Differentiated Services Test: Report
CAR: TCP burst tolerance (cont) Tiziana Ferrari Differentiated Services Test: Report
CISCO: CB-WFQ Scheduling mechanism to provide minimum bandwidth guarantees to classes • class definition: precedence or MF classification --> OK • bandwidth allocation: no starvation, no bandwidth consumption --> OK • traffic isolation: scenarios • UDP high priority + UDP best-effort • TCP high priority + UDP best-effort • TCP high priority + TCP best-effort UDP --> OK (always) TCP: inconsistency Tiziana Ferrari Differentiated Services Test: Report
CB-WFQ: set-up Tiziana Ferrari Differentiated Services Test: Report
CB-WFQ: traffic isolation of a single TCP stream Tiziana Ferrari Differentiated Services Test: Report
CB-WFQ: traffic isolation of multiple TCP streams Multiple TCP streams get the bandwidth share as defined by CB-WFQ! Tiziana Ferrari Differentiated Services Test: Report
IBM: Self Clocked Fair Queuing • Policy = (traffic profile, validity period, diffserv action) • diffserv action = (type of marking, queue type, bandwidth) • optimum traffic isolation (tests only with UDP) Premium: 163.8 Kbps guaranteed (8% PPP link bw) Assured: 819.2 Kbps (40% PPP bw) Tiziana Ferrari Differentiated Services Test: Report
IBM: EF policing Small TCP burst tolerance in a policer can completely starve a TCP stream. Token bucket depth is key parameter --> tuning need according to the rate at which traffic is policed Tiziana Ferrari Differentiated Services Test: Report
IBM: EF policing (cont) EF with TCP traffic, target rate = 163 Kbps Tiziana Ferrari Differentiated Services Test: Report
QoS measurement • Resource allocation monitoring • for resource allocation and network dimensioning • performance measurement • passive • active (invasive traffic) • for service validation • to understand the effect on end-to-end performance of buffering in one router or in a chain Tiziana Ferrari Differentiated Services Test: Report
Performance measurement • parameters of interest: • one-way delay, requirements: clock synchronisation • GPS based synchronisation • SmartBits (Netcom Systems) • NTP (Network Time Protocol) • NTP client / server hierarchy • one-way delay variation • packet loss • throughput • RTT Tiziana Ferrari Differentiated Services Test: Report
Services: examples • Virtual leased line: • point to point • one to many • capacity allocation on congested links (e.g. US links) • better-than-best-effort (qualitative definition) • delay bound and delay variation sensitive traffic classes • rate limiting of invasive traffic • ... Tiziana Ferrari Differentiated Services Test: Report
More information at • Test programme: • http://www.cnaf.infn.it/~ferrari/tfng/diffserv.html • more info and test results: • http://www.cnaf.infn.it/~ferrari/tfng/ds-test.html Tiziana Ferrari Differentiated Services Test: Report