Wu-Hsiao Hsu

Virtual network mapping algorithm in the cloud infrastructure

Network virtualization has received considerable attention recently because a Cloud Provider (CP) that is responsible for deploying a substrate network in the cloud infrastructure uses network virtualization to support multiple Virtual Network (VN) requests over the shared substrate network. However, mapping multiple VN requests with constraints on virtual nodes and virtual links into a shared substrate network presents a significant challenge, and is considered an NP-hard problem. In this paper, we propose a heuristic mapping algorithm that handles online VN requests. The node mapping algorithm selects a substrate node for mapping that satisfies both a virtual nodes resource requirement and its amount of requested bandwidth. The link mapping algorithm either maps a virtual link to the shortest substrate path that satisfies the requested bandwidth of the virtual link or uses the cut-shortest path approach to map a virtual link to multiple substrate paths that satisfy the requested bandwidth of the virtual link. The path migration algorithm migrates virtual links to different substrate paths to maximize the number of accepted VN requests in a substrate network. Simulation results show that the proposed heuristic mapping algorithm uses resources more efficiently, produces more revenue, and has better performance than existing mapping approaches.

DiffServ-based bandwidth-constrained anycast routing in a mobile IPv6 network

This paper proposes a differentiated service (DiffServ)-based bandwidth-constrained network layer anycast routing for a mobile IPv6 network. The proposed algorithm comprises three steps: (1) When moving into a new foreign network, the mobile node (MN) chooses a feasible path to easily locate the nearest server that can serve it. To perform path selection, a multi-level list structure (MLLS) is defined and used to store the available bandwidth and hop count of a path. (2) An ingress extended edge router (E-ER) uses the multi-level list algorithm (MLLA) to examine the MLLS for selecting the feasible path with the smallest number of hops that satisfies the MN bandwidth request. (3) The E-ER uses a condition of path change (CPC) variable to determine whether it should optimize the route between itself and the server. Simulation results indicate that the proposed algorithm offers better performance than the existing related approaches in terms of the average running time, request blocking rate, and reducing the control overhead of the simulated network. Copyright

Multiple path selection algorithm for DiffServ-aware MPLS traffic engineering

This paper proposes a new per-class bandwidth constraint algorithm, called the multipath selection algorithm (MSA), for a DiffServ-aware traffic engineering (DiffServ-TE). The MSA comprises three steps. First, a given source uses the MSA to find multiple label switch paths (LSPs) from the source to a destination for a specific class type (CT). Second, the source uses the available bandwidth of the CT on all the links along these LSPs to allocate the initial traffic to the selected LSPs. Third, the source dynamically adjusts traffic to these LSPs based on individual round trip time. Simulation results indicate that the proposed algorithm offers better performance than existing approaches in average transmission time, average packet loss rate, average throughput, and available bandwidth variance for each link.

Virtual Network Mapping through Path Splitting and Migration

Mapping multiple virtual network (VN) requests with constraints on virtual nodes and virtual links into a shared substrate network presents a significant challenge and is an NP-hard problem. In this paper, we propose a heuristic node and link mapping algorithm that handles online VN requests. The node mapping algorithm selects a substrate node that satisfies both a virtual nodes resource requirement and the amount of requested available bandwidth. The link mapping algorithm focuses on mapping a virtual link to a substrate path or paths. In addition, path migration is used to migrate virtual links to different substrate paths, which maximizes the number of coexisting VNs in a substrate network and increases the revenue of the Infrastructure Providers (InP). Simulation results indicate that the proposed algorithm offers better performance than existing mapping approaches.

An integrated end-to-end QoS anycast routing on DiffServ networks

Anycast is the point-to-point communication between a single client and the nearest destination server identified by an anycast address in traditional best-effort networks such as the Internet. Conversely, an anycast service in a QoS-based network is required to choose a path with bandwidth guarantees and to select a server with capacity guarantees. This paper proposes an approach which integrates both a network layer anycasting service and an application layer anycasting service to select a path and server that satisfies the client QoS constraint in DiffServ networks. In order to perform a path selection algorithm, the proposed protocol uses the idea of residual capacity to select all paths which match the client bandwidth requirement. In order to perform the server selection process, each server resource management (SRM) uses the server weight of each server to select the best server that has sufficient capacity for the client request. Furthermore, the design of multiple SRMs eliminates the single-point-of-failure problem and increases reliability in a DiffServ network. The simulation results have shown that the proposed protocol has much faster server response times and has the lowest request blocking rate compared to other existing protocols, especially in the case where the traffic load is high.

Smart handoff scheme for supporting roaming between WLAN and GPRS networks

Recent advancement in personal communication services (PCS) provides wireless multimedia services for users on the move. For establishing the convenient access environment, the evolution of IP-based integrated scheme becomes very urgent for public wireless access systems. In this paper, we clarify the well-known issues of integrating the general packet radio service (GPRS) network and IEEE 802.11 wireless local area network (WLAN). These issues include the inconsistence in transmission rate and coverage area, the difficulty in seamless handoff, the complexities of mobile IP and address translations, and so on. Based on classified cases, we propose the circular probe strategy (CPS) to measure the precise handoff latency of a mobile node roaming from one network to another. This information helps mobile nodes and its home agent (HA) to decide the appropriate handoff timing, to maximize the data transmission rate and to perform seamless handoff in the heterogeneous system.

Cross-layer end-to-end label switching protocol for WiMAX-MPLS heterogeneous networks

The integration of WiMAX networks and multi-protocol label switching (MPLS) networks, called WiMPLS networks, is the trend for nomadic Internet access in the fourth generation (4G) wireless networks. The base station (BS) in such heterogeneous networks will play the role of bridge and router between the IEEE 802.16 subscriber stations (SSs) and MPLS networks. However, there is no such integrated solution so far and the switching efficiency of the BS should be considered as well. This paper, therefore, adopts a cross-layer fashion (from network layer to MAC layer) to design the end-to-end label switching protocol (ELSP) for filling this gap. ELSP provides the mechanism of end-to-end (SS-to-SS) and layer 2 switching transfer for switching performance enhancement by assigning the SS with the MPLS labels (M-labels). The M-label can be carried by the IEEE 802.16e extended subheader within the MAC protocol data unit (MPDU), which is fully compliant with the IEEE 802.16e standard. The security issue caused by M-label usage is also concerned and solved in this paper. This paper also reveals an extra advantage that the switching delay of the BS achieved by ELSP can be as low as hardware-accelerated IP lookup mechanism, e.g., ternary content addressable memory (TCAM). Simulation results show that ELSP efficiently improves the end-to-end transfer delay as well as the throughput for WiMPLS heterogeneous networks.

DiffServ-aware multicasting in a mobile IPv6 network

In this paper we propose an algorithm to establish a DiffServ-Aware Multicast Tree (DAMT) in a mobile IPv6 network. We first use the modified Prim’s algorithm to construct the Maximum Available Bandwidth Map (MABM) in a mobile IPv6 network. Then, both the extended multicast routing table stored in each Edge Router (ER) and the Interface QoS Table (IQT) stored in each Core Router (CR) are created and used to construct a DAMT. Finally, the constructed MABM can be rearranged (R-MABM) in order to keep the same QoS as a Mobile Node (MN) had in the home network when the MN has moved to a new ER. The proposed algorithm solves the architectural conflicts for the integration of multicasting support in the Diffserv network as well as the problems of the Neglected Reservation Sub-tree (NRS) and receiver heterogeneity. It also solves a MN’s QoS requirement when the MN moves from one network to another. Simulation results indicate that the proposed algorithm offers better performance than existing approaches in average delay, average packet loss rate, average throughput, the number of control overhead, the usage of bandwidth, the amount of saved link bandwidth, and the accumulated time to find a feasible path, especially in the case where the number of new join MN is large and traffic loads are heavy.

Simple path diversity algorithm for interdomain routing

The border gateway protocol (BGP) handles the task of establishing paths between autonomous systems (AS). As the BGP does not incorporate measures of round-trip time into the decision and does not provide path diversity for a destination, this study proposes a simple path diversity algorithm (SPDA) for interdomain routing. The SPDA comprises three steps. First, the BGP is extended to allow the BGP routing table to save multiple paths for any destination. Second, a given source node determines whether the point of congestion occurs along the best path. Third, if congestion occurs along the best path, the source node specifies an alternate path and directs traffic over the best and alternate paths concurrently. Simulation results indicate that the SPDA produces better performance than existing approaches in average end-to-end delay, average packet loss rate and average throughput.

A Simple Path Diversity Algorithm for Interdomain Routing

This paper proposes a simple path diversity algorithm (SPDA) for Interdomain routing. The SPDA uses source routing to find an alternate path from a source to a destination, and allows flexible division of traffic over the best and alternate path. The SPDA comprises three steps. First, the BGP protocol is extended to allow the BGP routing table to save the multiple paths for any destination. Second, a given source is able to detect whether the point of congestion along the best path occurs. Third, if congestion occurs, the source will specify an alternate path and direct traffic over the best and alternate path concurrently. Simulation results indicate that the SPDA produces better performance than existing approaches in average end-to-end delay.