Multiple Layer Traffic Engineering in NTT Network Service

Transcription

1 Multi-layer traffic engineering in photonic-gmpls-router networks Naoaki Yamanaka, Masaru Katayama, Kohei Shiomoto, Eiji Oki and Nobuaki Matsuura * NTT Network Innovation Laboratories * NTT Network Service Systems Laboratories 3-9- Midori-cho, Musashino-shi, Tokyo, Japan Abstract-This paper describes multi-layer traffic engineering and signaling technologies in photonic-gmpls-router networks. Multi-layer traffic engineering, which yields the dynamic cooperation of IP and photonic layers for providing IP services -effectively, is described. To realize multi-layer traffic engineering, we propose an OSPF extension that advertises both the number of total waves and the number of unused waves, and an -TE extension that minimizes the number of wave conversions needed. In addition, this paper presents a heuristic-based multi-layer topology design scheme that uses IP traffic measurements in a generalized multi-protocol label switch (GMPLS). Our design scheme yields the optical label switch path (OLSP) network topology, i.e. OLSP placement, that minimizes network, in response to fluctuations in IP traffic demand. In other words, the OLSP network topology is dynamically reconfigured to match IP traffic demand. Networks are reconfigured by the proposed scheme so as to utilize the network resources in the most -effective manner. I. INTRODUCTION The explosion of Internet traffic has strengthened the need for high-speed backbone networks. The growth in Internet-protocol (IP) traffic exceeds that of IP packet processing capability. Therefore, the next-generation backbone networks should consist of IP routers with IP-packet switching capability and optical cross-connects (OXC) with wave-path switching capability to reduce the burden of heavy IP-packet switching loads. In addition, IP traffic fluctuates often over periods as short as hours. These needs are satisfied by the photonic GMPLS router developed by NTT []; it offers both IP packet switching and wave-path switching capabilities [][3]. Wave paths, called optical label switch paths (OLSP) are set and released in a distributed manner based on the functions offered by the generalized multi-protocol label switch (GMPLS). Since the photonic MPLS router offers the functions of both switching and GMPLS, it enables us to create the optimum network configuration by considering IP and photonic network resources in a distributed manner. An OSPF extension and a very sophisticated signaling mechanism using an -TE extension have been proposed and are going to be implemented. The OSPF extension is to advertise both the total number of waves and the number of unused waves for each link. This information is passed via OSPF packets to all egress edge nodes so that they are aware of the GMPLS link state. Based on the least-load rule, egress edge nodes try to establish new OLSPs by using the -TE extension signaling protocol. The signaling packet selects the proper wave that uses the least number of wave converters because wave conversion is a very expensive operation. In addition, multi-layer traffic engineering, which ensures that IP and photonic layers dynamically cooperate, is required to provide IP services -effectively. This paper proposes a heuristic-based multi-layer topology design scheme that uses IP traffic measurements. The proposed scheme yields the optimum OLSP network topology, i.e. OLSP placement, so as to minimize network, in response to fluctuations in IP traffic demand. In other words, OLSP network topology is dynamically reconfigured to match IP traffic demand. II. GMPLS-BASED PHOTONIC MULTILAYER ROUTER ARCHITECTURE The structure of the photonic GMPLS multi-layer router is shown in Fig.. It consists of an IP router, wave router, and photonic-gmpls-router manager. The router is based on a photonic universal platform with the addition of 3R (Retiming, Reshaping and Regeneration) functions, wave conversion, and layer-3 functions. These functions are used adaptively. In other words, if the signal is degraded by fiber loss as well as nonlinear effects such as PMD (polarization mode dispersion) or ASE (Amplified Spontaneous Emission), the 3R function is activated. In addition, wave conversion is also used when signaling is blocked by wave overbooking. Of course, layer 3 (L3) packet forwarding is adaptively used when the L3 packet forwarding is performed. Note that the photonic-gmpls router can transfer 3 types of signal as shown in Fig.. -A is transparent transfer or bit-rate restriction free. - are λ relay connections that need wave conversion. - connections are also supported by the adaptive use of the 3R function. Finally, type-c involves layer 3 switching functions for operations such as MPLS path aggregation and L3 packet level forwarding. The photonic-gmpls router can support all transfer capabilities. In the photonic-mpls-router manager, the GMPLS controller distributes own IP and photonic link states, and collects the link states of other photonic MPLS routers. IP traffic is always monitored, and the captured data is passed to the GMPLS controller through a low-pass filter [4]. A multi-layer topology design algorithm processes the collected IP and photonic link states and the collected traffic data.

2 -C -B -A WDM DMUX Optical Path Optical SW Monitor λ-conv. 3R L-Trunk IP/MPLS router# FE FE S w Router# L-SW # Monitor Multicast L/L3-Trunk Photonic IP link IP traffic monitor link state state with low-pass filter Multi-layer topology design algorithm OSPF extension -TE extension GMPLS controller Fig. Photonic GMPLS router manager Hardware structure -A (Layer 3 Packet forwarding) -B (Optical 3R or wave conversion path) -C (Optical transparent path, Bit-rate restriction free) Hardware manager Software structure LRU=WDM+Optical SW+L-Trunk LRU : Lambda Routing Unit FE : Forwarding Engine SW : Switch λ-conv : Lambda converter NE : Network Element Structure of photonic MPLS router with multi-layer traffic engineering based on IP traffic monitoring III. GMPLS MULTI-LAYER NETWORK CONTROL AND A TOPOLOGY DESIGN SCHEME A. Framework The bandwidth granularity of the photonic layer is coarse and equal to wave bandwidth, λ, i.e..5gbit/s or Gbit/s. On the other hand, the granularity of the IP layer is flexible and well engineered [5]. When traffic demand between source and destination IP routers is much less than λ, the cut-through wave path between the source-destination IP routers is not fully utilized and so is not -effective. In this case, several IP traffic streams should be merged at some IP transit routers to better utilize the wave bandwidth at the of IP-packet processing at the transit nodes. On the other hand, when traffic demand between source and destination IP routers approaches or exceeds λ, such stream multiplexing is dropped. Thus, the setting of cut-through paths between source-destination IP routers depends on IP traffic, and the OLSP topology of the photonic layer should change dynamically according to fluctuations in IP-traffic demand to optimize network resource utilization. For this purpose, our objective is to minimize network Z, which is formulated as follows. z = C + node Clink = α + β () ip ijpk, i r i j p k where r ip is the of port p in router i, l ijpk is the of wave path k at port p in fiber link i j. α/β is the Node/link ratio. α/β is set to more than one, because IP routers have IP-packet processing functions such as table look up and packet-based switching in addition to wave routing functions. l To minimize Z, we adopt the extended version of the BXCQ (branch exchange with quality-of-service constraints) scheme presented in [6], named EBXCQ. The BXCQ scheme was originally intended for multi-layer ATM network design. In BXCQ, the addition/elimination of links is iterated to solve a topological optimization problem with quality-of-service constraints, such as delay and blocking probability. In EBXCQ, the number of ports in both IP routers and wave routers, as well as the number of waves per fiber are also considered as constraints in addition to the constraints considered in BXCQ. B. Numerical results We have confirmed the effectiveness of EBXBQ using a LATA network model, see Fig. [6]. Fig. also shows that the optimum OLSP topology changes with the IP traffic demand between source-destination IP routers. We assume that IP traffic demands between source-destination IP routers are evenly distributed and that each wave is converted at each photonic MPLS router. Wave bandwidth is set to.5gbit/s. We used the metric of the average node degree, D, to characterize and evaluate OLSP network topologies. D is the average number of other IP routers to which individual IP routers are connected by OLSPs. As traffic demand increases, the optimum OLSP topology becomes a mesh because each OLSP bandwidth becomes effectively utilized. Changing the OLSP topology in response to traffic demand fluctuations dramatically reduces network. An estimation of the network reduction possible is shown in Fig. 3. Network is normalized by the fixed optimum topology T=3 [Mbit/s]. If the OLSP network topology is changed based on traffic measurements; the reduction effect can exceed 5%.

3 Average node degree D Traffic demand between S-D pairs, T [Mbit/s] Fig. Numerical model and optimum OLSP topology determined by IP traffic demand Normalized network Calculation model α/β= Physical link Fixed topology Changing topology Network reduction. α/β= Fig. 3 Traffic demand between S-D pairs, T [Mbit/s] Cost reduction effect of reconfigurable multi-layer network. IV. ROUTING AND SIGNALING EXTENSION FOR PROPOSED PHOTONIC GMPLS Optical wave conversion is very ly and needs a lot of hardware. In other words, the transparent bit-rate restriction free network is very attractive from the viewpoints of low and high flexibility. Accordingly, our OSPF extension and -TE extension can provide effective route selection as well as the least number of wave conversions [7][8]. First, for route advertisement, the OSPF extension is used as shown in Fig. 4. Each photonic GMPLS router advertises its total number of used and unused waves. According to this advertisement, edge nodes can discern GMPLS link state and can thus select the least expensive path. In addition, we propose to extend the sub-tlv of IETF specification (draft-ietf-ccamp-ospf-gmpls-extensions-.txt). The extensions include 3R resource information and statistical information such as utilization of each wave, and 3R and wave conversion resources. This information will be used for source routing based on a combination of shortest path first and load information. For the signaling function, we offer the -TE extension. Link by link routing is used based on source routing. The first photonic GMPLS router sets the unused wave information using bit map format in signaling. Each transit photonic GMPLS router over-writes this information by placing And between arriving signaling unused wave bitmap and its own unused waves. If there is no unused wave, wave conversion is used. The router, which offers wave conversion, creates a new unused wave bitmap and sends it to the next router. [9] This signaling and routing technique minimizes the GMPLS link state. OSPF extension OPSF extension OPSF extension Advertisement A B C : Number of waves : Number of unreserved waves WC: Wave Converter Signaling -TE extension Route A-B-C -TE extension : Unused wave : Used wave Unused waves for both link A and B WC If no unused waves match then use wave conversion wc Fig. 4 Photonic GMPLS extensions for both routing and signaling based on OSPF and -TE

4 frequency of wave conversion in the network and so can provide very effective photonic networks. V. ADAPTIVE ROUTING TECHNIQUE IN PHOTONIC NETWORK Wave conversion is adaptively used in the network. We propose three types of routing method in this paper. (Least loaded) routing is popular and selects the least loaded wave resources route. SPF-WC (Shortest Path Finding with Wave Conversion) is based on the SPF algorithm, but wave conversion is adaptively used if wave blocking occurs. The third algorithm is AWCC (Average of Wave Conversion Capability). AWCC selects the best combination of link and wave conversion. Fig. 5 shows an example of a network model. / is wave load. If the number is small, there are few empty waves for the link. Least loaded () with hop number limit selects the least loaded link as shown in Fig. 5. On the other hand, SPF-WC uses wave conversion under the shortest path finding algorithm. The algorithm needs a lot of wave conversions but it does select the shortest path. Fig. 6 shows the AWCC algorithm; Node A and B in the figure have wave conversion capability, however, node A has few wave conversion resources. So node A s conversion is 5 while that of node B is only. AWCC selects the least route considering hop number and wave conversion. Table compares the routing algorithms. Fig. 7 shows calculation s for networks with average node degree of 8. The results show AWCC needs a lot of calculation power. Fig. 8 shows network efficiency using the proposed algorithms. The results show tends to create new OLSPs because each wave usage becomes on average. We are now evaluating the trial network using these algorithms. Table Comparison of routing algorithms Algorithm Resource efficiency Calculation time SPF-WC AWCC 3 : best 3: worst 3 Source SPF-WC Destination...9 LL.9.8 Fig. 5 Routing examples of and SPF-WC routing methods Calculation Time [msec] 5 5 AWCC 5 5 Node Number Fig. 7 Calculation power of proposed OLSP routing method Route- B = A = 5 C λ-plane λ-plane Number of OLSPs 5 5 SPF-WC Route- λ3-plane 5 Carried Load [Gb/s] Fig. 6 Routing example of AWCC routing method Fig. 8 Network efficiency of proposed algorithms

5 VI. CONCLUSIONS This paper presented multi-layer traffic engineering, signaling and routing technologies for photonic-gmpls-router networks. The photonic-gmpls-router network architecture is based on OSPF and -TE extensions. The proposed routing and signaling techniques minimize the number of wave conversions needed and so yield very effective photonic networks. This paper also described a heuristic-based multi-layer topology design scheme, called EBXCQ, for IP photonic networks. By monitoring IP traffic loads, photonic MPLS routers are controlled to dynamically change the network configuration; network resources are utilized efficiently in a distributed manner. Finally an optimized routing method was introduced. High-performance and effective photonic networks can be realized using the proposed technologies. REFERENCES [] Y. Yamabayashi et al., Autonomously Controlled Multiprotocol Wave Switching Network for Internet Backbones, IEICE Trans. Commun., vol. E83-B, no., pp. -5,. [] K. Shimano et al., MPLambdaS Demonstration Employing Photonic Routers (56X56 OLSPS) To Integrate Optical and IP Networks, National Fiber Optic Engineers Conf. Tech.Proc. p. 5. [3] F. Kano et al., Development of Photonic Router (56X56 OLSPS) Realize Next Generation Network with MPLambsaS Control, National Fiber Optic Engineers Conf. Tech. Proc. p [4] K. Shiomoto, A Simple Bandwidth Management Strategy Based on Measurements of Instantaneous Virtual Path Uti-lization in ATM Networks, IEEE Trans. Networking, vol. 6, no. 5, pp , 998. [5] K. Shimano et al., MPLambdaS Demonstration Employing Photonic Routers (56X56 OLSPS) To Integrate Optical and IP Networks, National Fiber Optic Engineers Conf. Tech.Proc. p. 5. [6] E. Oki and N. Yamanaka, Impact of Multimedia Traffic Characteristic on ATM Network Configuration, J. Network Systems Management, vol. 6, no. 4, pp , 998. [7] W. Imajuku et al., Multi-layer routing using multiple-layer switch capable LSRs, IETF draft, draft-imajuku-ml-routing-.txt, February. (working in progress) [8] E. Oki et al., Requirements of optical link-state information for traffic engineering, IETF draft, draft-oki-ipo-optlink-req-.txt, February. (working in progress) [9] N. Matsuura et al., Requirements for using -TE in GMPLS, IETF draft, draft-matsuura-gmpls-rsvp-requirements-.txt, February. (working in progress)