Stacked-VLAN-Based Modeling of Hybrid ISP Tra ffi c Control Schemes and Service Plans Exploiting Excess Bandwidth in Shared Access Networks Dr Kyeong Soo (Joseph) Kim Department of Electrical and Electronic Engineering Xi’an Jiaotong-Liverpool University 15-16 September 2016 1 / 21
Outline Introduction Review of Hybrid ISP Tra ffi c Control for Shared Access Modeling of Hybrid ISP Tra ffi c Control Schemes and Service Plans based on Stacked VLANs Summary 2 / 21
Next . . . Introduction Review of Hybrid ISP Tra ffi c Control for Shared Access Modeling of Hybrid ISP Tra ffi c Control Schemes and Service Plans based on Stacked VLANs Summary 3 / 21
Current ISP Tra ffi c Control Architecture Access Network UNI SNI R D R U Subscriber Subscriber Unit 1 1 R F R B R B Shared App. Access (= R D ) … … Router (O)DN * Server Switch R D R U Subscriber Subscriber Unit N N Access Switch I/F with Relay Unit SNI (O)DN ISP Per-Subscriber Port Port MAC MAC Filter Traffic Filter Control TBF * … Egress Scheduler * (O)DN: (Optical) Distribution Network Classifier * SNI: Service Node Interface * TBF: Token Bucket Filter TBF * UNI: User-Network Interface 4 / 21
Issues with Current ISP Tra ffi c Control The arrangement of tra ffi c shapers and a scheduler prevents subscribers from sharing available bandwidth. ��� � � ������ ��������� ���������� ��� 5 / 21
Toward Fully-Shared Access To implement fully-shared access networks for better resource utilization and higher energy e ffi ciency, we have been studying the following: ISP tra ffi c control schemes enabling excess bandwidth allocation 12 . Hybrid ISP tra ffi c control architecture and service plans exploiting excess bandwidth 3 . Implementation of simulation models for the proposed tra ffi c control architecture and service plans ( reported in this paper ). 1 K. S. Kim, IEEE Commun. Lett. , vol. 18, no. 4, Apr. 2014. 2 K. S. Kim, Proc. IEEE Sarno ff Symposium 2015 , Aug. 2015. 3 K. S. Kim, Proc. ICTC 2014 , Busan, Korea, Oct. 2014. 6 / 21
Next . . . Introduction Review of Hybrid ISP Tra ffi c Control for Shared Access Modeling of Hybrid ISP Tra ffi c Control Schemes and Service Plans based on Stacked VLANs Summary 7 / 21
Hybrid ISP Tra ffi c Control Architecture I TBF … TBF Egress Classifier Virtual Subscriber Relay Regular Output TBM * Unit Scheduler Port Scheduler … Allocating TBF Excess Bandwidth TBM * TBM: Token Bucket Meter 8 / 21
Hybrid ISP Tra ffi c Control Architecture II Backward compatibility • A group of new service plan subscribers are treated as one virtual subscriber under the existing flat-rate service plan. Better resource utilization • The new service plan subscribers can fully share the bandwidth within the group. Gradual introduction • Migration to a fully-shared access will be completed when all the subscribers of the existing service plan move to the new one. 9 / 21
Next . . . Introduction Review of Hybrid ISP Tra ffi c Control for Shared Access Modeling of Hybrid ISP Tra ffi c Control Schemes and Service Plans based on Stacked VLANs Summary 10 / 21
Modeling of Hybrid Tra ffi c Control I Major requirements: Backward compatibility with the existing VLAN implementations in INET-HNRL, including • EthernetFrameWithVLAN message format • MACRelayUnitNPWithVLAN and VLANTagger modules Expandability to stack more than two VLAN tags 11 / 21
Modeling of Hybrid Tra ffi c Control II Two possible approaches: Integrated approach • The whole scheduling is implemented as one integrated scheduler like the hierarchical token bucket (HTB) scheduler 4 . Modular approach • Separate schedulers (e.g., a scheduler based on TBF shaping and a DRR-based scheduler enabling excess bandwidth allocation 5 ) are combined into one. 4 M. Devera. Linux HTB home page. 5 K. S. Kim, Proc. IEEE Sarno ff Symposium 2015 , Aug. 2015. 12 / 21
Identification of Subscriber Frames I For the existing tra ffi c control scheme, the whole frames from the subscribers of the hybrid scheme are identified and treated as a group (i.e., ��� one virtual subscriber ). � For the new ��� excess-bandwidth-allocating ������������������ ������� tra ffi c control scheme, the ��� � ��������� frames from each subscriber ��������� � ���������� are identified and treated as a ��� ������ ��������� separate flow. ��� This can be done by IEEE 802.1Q stacked VLANs (also called provider bridging and Q-in-Q ). 13 / 21
Frame Formats for VLAN Stacking Ethernet II Frame Destination Source Frame Check Address Address Ethertype Payload Sequence (DA) (DA) (FCS) Tag Ethernet II Frame with Single VLAN TAG Push Tag Destination Source Frame Check 802.1Q Header Address Address Ethertype Payload Sequence Pop (TPID=0x8100) (DA) (DA) (FCS) Ethernet II Frame with Double VLAN TAGs Destination Source Frame Check 802.1Q Header 802.1Q Header Address Address Ethertype Payload Sequence (TPID=0x88A8) (TPID=0x8100) (DA) (DA) (FCS) S-VLAN C-VLAN 14 / 21
Ethernet II Frame Message Definitions Without VLAN stacking: packet EthernetIIFrameWithVLAN extends EthernetIIFrame { uint16_t tpid = 0x8100; // tag protocol identifier (16 bits; set to 0x8100) uint8_t pcp; // priority code point for IEEE 802.1p class of service (3 bits; 0 (lowest) to 7 (highest)) bool dei; // drop eligible indicator (1 bit) uint16_t vid; // VLAN identifier (12 bits; 0x000 and 0xFFF are reserved, which allows up to 4094 VLANs) } With VLAN stacking: cplusplus {{ #include <stack> #include "VLAN.h" // define VLANTag struct typedef std::stack<VLANTag> VLANTagStack; }} class noncobject VLANTagStack; packet EthernetIIFrameWithVLAN extends EthernetIIFrame { uint16_t tpid; // tag protocol identifier (16 bits; set to 0x8100 for C-TAG & 0x88A8 for S-TAG) uint8_t pcp; // priority code point for IEEE 802.1p class of service (3 bits; 0 (lowest) to 7 (highest)) bool dei; // drop eligible indicator (1 bit) uint16_t vid; // VLAN identifier (12 bits; 0x000 and 0xFFF are reserved, which allows up to 4094 VLANs) // optional; IEEE 802.1Q-in-Q stacked VLANs. VLANTagStack innerTags; // based on std::stack } 15 / 21
Stacked-VLAN-Based Modeling of An Access Network with Hybrid ISP Tra ffi c Control ���������������������������� ����� �������������������������������� ���������������������������� ������ ������������������������ ������� 16 / 21
Modules with VLAN Support 17 / 21
Backpressure between Schedulers Two schedulers are currently connected based on Ethernet flow control, which halts the transmission of the whole frames. For better control, we need an extension (like IEEE 802.1Qbb) to halt the transmission of non-conformant frames only . �������� �������� ��� ��� ������������������ ��� � � ��������� � ���������� ������� ������ ������ ��������� ���� ��������� ��� ��� ��� ��� ��� ������������ ������������� 18 / 21
Throughput Example RR+TBF (Original) DRR+TBM (Proposed) CSFQ+TBM 100 90 1.11s Throughput [Mb/s] 80 70 60 50 40 30 20 10 0 9 10 11 12 13 14 9 10 11 12 13 14 9 10 11 12 13 14 Time [s] 19 / 21
Next . . . Introduction Review of Hybrid ISP Tra ffi c Control for Shared Access Modeling of Hybrid ISP Tra ffi c Control Schemes and Service Plans based on Stacked VLANs Summary 20 / 21
Summary Discussed the issues in current practice of ISP tra ffi c shaping and related flat-rate service plans in shared access networks. Reviewed alternative service plans based on new hybrid ISP tra ffi c control schemes exploiting excess bandwidth. Reported the current status of our modeling of the hybrid ISP tra ffi c control schemes and service plans with OMNeT ++/ INET-HNRL based on stacked VLANs. 21 / 21
Recommend
More recommend
