Service differentiation for variable length packets in OPS with recirculating FDLs Chris Develder , Jan Cheyns Mario Pickavet, Piet Demeester Dept. of Information Technology (INTEC) Ghent University - IMEC, Belgium UNIVERSITEIT GENT
Outline • Context: the DAVID-project • Switch architecture • QoS approaches • Performance criteria • Simulation set-up • Results - influence of nr. of buffer ports - influence of “class offset” - influence of buffer delay - influence of load • Conclusions PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 2
Context • Optical packet switching • Network - MAN: metro rings, MAC protocol http://david.com.dtu.dk - WAN: full-mesh, (G)MPLS-based control • Key components - Ring node (OPADM), Hub, Gateway, OPR packet WAN router hub hub gateway gateway ring ring node node MAN IP router IP router PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 3
Switch Architecture • Node in core OPS network (backbone) • Switch functionality: - fully non-blocking switching matrix (SOA-based) - wavelength conversion to solve contention - FDLs to provide buffering F fibers all-optical W wavelengths space switch B buffer ports FDL delay = D PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 4
QoS approaches (1) • Known approaches to provide QoS in OPS/OBS: - resource reservation (dedicated wavelength converters, buffers); static or dynamically - OBS-JET with differentiated offsets - burst segmentation: tails have larger survival chances (implicit QoS) - intentional drops: drop low priority traffic to free resources… tail tail OBS-JET: higher offset = higher priority segmentation: drop head of overlapping packet • This study: - no resource reservation, no intentional drops, no segmentation PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 5
QoS approaches (2) • Header offset differentiation: offset O • well-known OBS-JET; high priority = larger offset • high priority bursts are known to the switch longer in advance • Look-ahead: look-ahead delay H • no different offsets; but look-ahead delay at input: time H to “change our mind” • Slotted control: slot time T • headers are delayed electronically, and handled in batches each timeslot PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 6
Scheduling algorithm • for each arriving packet, do: - check if there is a free wavelength on the output port it’s destined for, using LAUC-VF - if no free wavelength: find free buffer wavelength, using LAUC - if packet is buffered: do not reserve output wavelength yet, but repeat scheduling upon re-entrance of switch (PostRes) • look-ahead: - high priority packets may preempt low priority ones - preempted packets are re-scheduled using same algorithm upon time of preemption • slotted control: - packets scheduled at same time are sorted: first the high priority packets PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 7
Performance criteria • loss rate: - amount of data lost / amount of data sent • fairness: - are longer packets strongly discriminated against? PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 8
Simulation set-up • Parameters: W wavelengths - F=6 input/output fibres - W=8 wavelengths per i/o fibre F fibers all-optical - B=0..64 recirculating buffer ports space switch - D= delay in buffer B buffer ports - L= average packet length FDL delay = D - 40% high priority; 60% low priority • Traffic model: - train length: minimal L/2, average L - train length distribution: (length - L/2) follows neg. expo distribution - train inter-arrival: Poisson process - uniform distribution over output fibres PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 9
Influence of nr. of buffer ports (1) • settings: - O=H=T=2L; buffer D=2L 1.E+00 obs, total obs, high 1.E-01 - load 0.8 obs, low 1.E-02 loss rate lah, total 1.E-03 lah, high 1.E-04 • loss rates: lah, low 1.E-05 slot, total - fairly strong class separation 1.E-06 slot, high 0 16 32 48 64 - slotted control higher avg nr. buffer ports B slot, low loss and less differentiated - no significant difference obs, total 2 between look-ahead and avg. nr. recircs recv obs, high 1.5 diff. offsets obs, low low priority lah, total 1 • delays: lah, high overall avg 0.5 lah, low - slotted worse than others high priority slot, total 0 - more buff ports = delay slot, high 0 16 32 48 64 nr. buffer ports B slot, low instead of loss PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 10
Influence of nr. of buffer ports (2) • unfairness: - more pronounced for look-ahead; which - stems from preemption: re-scheduling packets leads to less optimal wavelength allocation (ie. worse config of “gaps” left for other packets) 0.14 0.12 loss rate 0.10 off, B=8, total 0.08 lah, B=8, total slot, B=8, total 0.06 0.5 1 1.5 2 2.5 3 3.5 burst size (unit L) packet size (unit L) PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 11
Influence of class offset • settings: - buffer B=8, D=4L; 0.12 - load 0.8 off 0.11 loss rate - varying “class offset” O=H=T 0.10 lah 0.09 0.08 slot 0.07 • overall loss rate: 0.06 0 1 2 3 4 class-offset (unit L) - stronger class separation: high priority forces more low priority losses 1.15 avg. pktsize drop - boundary (reached for smaller off 1.10 “class offsets” with look-ahead) lah 1.05 • unfairness: 1.00 slot 0.95 - significantly stronger for look- 0 1 2 3 4 class-offset (unit L) ahead PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 12
Influence of load 1.E+00 obs, pri.1 obs, pri.0 lookahead, pri.1 1.E-02 lookahead, pri.0 B=0 slottedctl, pri.1 loss rate slottedctl, pri.0 1.E-04 obs, pri.1 obs, pri.0 lookahead, pri.1 1.E-06 lookahead, pri.0 B=8 slottedctl, pri.1 slottedctl, pri.0 1.E-08 0 0.2 0.4 0.6 0.8 1 load • settings: • load =0.1…0.9; buffer: B=0 or 8, D=2L; “class offset”: O=H=T=2L • loss rates: • class separation slightly diminishes for increasing load • slotted control achieves much weaker separation; esp. if there is buffer (B=8) • buffer very much helps to increase sustainable load PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 13
Conclusions • compared various QoS approaches • slotted control achieves less strong separation, but - simpler scheduling algorithm - may be suitable for low to medium loads (cf. for load=0.5, high priority loss ~1E-6) • look-ahead - achieves equally good (or slightly better) loss rates and delays as OBS-JET with differentiated offsets - separation limit is reached for shorter “class offsets” than OBS-JET - induces more intra-class unfairness (i.e. stronger discrimination of longer bursts) PS.Mo.C8, 29 Sep. 2003 C. Develder, et al., "Service differentiation for var. length packets in OPS with recirculating FDLs" 14
That’s all, folks! … thanks for your attention … any questions? UNIVERSITEIT GENT
Recommend
More recommend
