Dynamic Path Bandwidth Allocation for 100010-Scale Optical Layer-2 Switch Network based on Hierarchical Timeslot Allocation Algorithm and Timeslot Converter

Abstract

We are developing optical layer-2 switch network that achieves dynamic path bandwidth allocation (DPBA) for efficient aggregation in metro NWs. We show the experimental results of DPBA cycle according to variations in traffic on NW scale of 1000x10. IntroductionNetwork (NW) traffic is increasing exponentially, and it is therefore becoming increasingly necessary to develop cost-effective NWs in terms of equipment cost and power consumption. For this trend, we are developing an optical layer-2 switch network 1 (OL2SW-NW) that can efficiently aggregate traffic in a largescale metro NW. The OL2SW-NW is based on a WDM/TDM ring NW of L WDM channels. This allows the bandwidth of wavelengths in the NW to be shared with the ground paths between 1000 aggregation switches (SWs) on the access NW and 10 IP routers (RRs) on the core NW. The OL2SW-NW enables NW bandwidth to be shared effectively as a TDM-PON 2 does by allocating timeslots (TSs) to each ground path and performing ADD/DROP of data according to allocated TSs (Fig. 1). The OL2SW-NW can dynamically change each bandwidth of ground path by changing the number of allocated TSs according to the amount of traffic on each ground path at every fixed time. To achieve the bandwidth allocation, traffic information on all ground paths is collected at a central TS scheduler (SCH) for TS allocation (TSA). Hence, collecting traffic information from all ground paths and TSA processing at the SCH will produce bottlenecks in large-scale NWs. In this paper, we show the experimental results of dynamic path bandwidth allocation (DPBA) in 1000x10-scale NW by using a novel TSA algorism and TS converters (TSCs).

Bufferless Bidirectional Multi-Ring Networks with Sharing an Optical Burst Mode Transceiver for Any Route

“…It can characteristically eliminate the L2SW located in the internal network, which is required for conventional aggregation networks, by aggregating the packet buffer to the entity of the edge node in the OL2SW-NW. This entity is called an "Ether-burst converter (EBC) [13]," and it converts variable-length 10G Ethernet (10GE) frames to TSs with a fixed length and transmits them according to the allocated TSs at a constant TDM frame (the length of TDM frame: L FR ). Each node deploys a WDM/TDM SW [12], which enables TS add/drop for any channel and any TS.…”

“…Each node deploys a WDM/TDM SW [12], which enables TS add/drop for any channel and any TS. To allocate the bandwidth for logical paths between EBCs, the OL2SW-NW has a TS scheduler that collects the traffic information on all logical paths for TSA calculated by a high speed TSA algorithm [13]. Then, the TS scheduler periodically allocates bandwidth at a constant cycle, T , which is an integral multiple of L FR for achieving DBA for large-scale aggregation networks [18].…”

“…The OL2SW-NW has to both change the route of TS transmission when the link failure occurs and control the DBA. Therefore, it is desirable for the achievement of both the fast restore from the occurrence of the failure and shortened DBA cycle to increase the efficiency of the accommodated traffic [13]. The control information related to DBA and the detection of a link failure is transmitted using control TSs between each node.…”

“…The control information related to DBA and the detection of a link failure is transmitted using control TSs between each node. As an example, the size of control information for DBA becomes about 10 Kbytes in the assumed network model according to [13]. If the interface speed for the control plane is 1 Gbps, the required size of a control TS become about 100 μs.…”

Evaluation of parallel processing control of virtual switch architecture on large-scale network

“…Our aim was to apply our virtual L2 switch architecture to a large-scale metro NW, assuming that the metro NW has 10 edge routers with 1000 FSs on the access side and 10 FSs on the edge router side [13] and aggregates two million flows. We used 30, 60, 90, and 120 DFCs because 15 prototype DFCs, which are implemented in low-end servers, can accommodate 100 K flows [8]; thus, we believe the number of DFCs can be less than several hundred.…”