Shuo-Cheng Hu

Routing interval: a new concept for IP lookups

Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP packet forwarding, which complicates the IP lookup. Currently, this process is mainly done in software and several schemes have been proposed for hardware implementation. Since the IP lookup performance is a major design issue for the new generation routers, in this article, we investigate the properties of the routing table and present a new approach for IP lookups; our approach does not require the rule of BMP and thus significantly reduces the complexity. By applying our proposed approach, the computation cost of existing schemes can be significantly reduced. We also propose an efficient IP lookup algorithm, with which we improve the binary search on prefixes to 30 MPPS (million packets per second) and 5,000 route updates/sec under the same experimental setup with an even larger routing table.

Hardware-based IP routing lookup with incremental update

Nowadays, the commonly used table lookup scheme for IP routing is based on the so-called classless interdomain routing (CIDR). With CIDR, routers must find out the best matching prefix (BMP) for IP packet forwarding, which complicates the IP lookup. Since the IP lookup performance is a major design issue for the new generation routers, we investigate the properties of the routing table and present a new IP lookup scheme. By adopting the extreme compression, the size of the forwarding table can be compressed to 360 Kbytes for a large routing table with 58,000 routing entries. It can accomplish one IPv4 route lookup within maximum nine memory accesses. Consequently, the data structure for the incremental update is introduced. With the new data structure, the proposed scheme can accomplish one route update within 500 ns. Even under the circumstance with 4000 route updates per second, the performance degradation is as low as 0.2 %.

Least required bandwidth in VP-based ATM networks

In an ATM network, a virtual path (VP) is a labeled path which can be used to transport a bundle of virtual connections (VCs) and to manage the resources used by these connections. The ATM multimedia leased line service is provided across user-to-user VPs. The customers can increase/decrease the VP bandwidth according to their demand. This paper proposes a method to derive the least required bandwidth on a VP which is the minimum amount of bandwidth estimated to satisfy the QoS requirement of the VCs. The least required bandwidth can be a warranty of the VP resource reallocation.

Virtual buffering strategy for GFR services in IP/ATM internetworks

We investigate the TCP packet transmission over the ATM guaranteed frame rate (GFR) service using the proposed virtual buffering strategy (VBS) which provides both minimum cell rate (MCR) guarantees and fair sharing to GFR VCs on a FIFO queue. VBS is composed of an emulation of per-VC queueing with weighted round robin (EWRR) and the differential packet discarding (DPD) buffer management scheme. From the simulation results, we demonstrate that VBS fulfils the requirements of GFR service as well as improves the TCP throughput.

Traffic control through in-service monitoring for QoS improvement in ATM networks

Asynchronous transfer mode (ATM) networks are expected to support multimedia communications with diverse traffic characteristics, these traffic are statistically multiplexed to efficiently utilize the network bandwidth. However, the high bandwidth utilization may deteriorate traffic-dependent quality of service (QoS). To accommodate QoS requirements while still maintaining high bandwidth utilization, it is necessary to prioritize users service requirements through effective resource management. This article presents a traffic control approach through in-service monitoring, which employs dynamic buffer allocation to allow complete buffer sharing by all service classes as well as improves QoS of delay-sensitive traffic using the so-called Earliest Expected Due-time First scheduling. The simulation results demonstrate that diverse multimedia traffic achieves satisfactory performance using our scheme. A feasible implementation architecture is also addressed.

Frame-based priority scheduling in hybrid IP/ATM networks

The exploding growth of World Wide Web applications in recent years has overwhelmed the internet with multimedia traffic. A future performance concern is to keep pace with the quickly growing bandwidth requirements. IP over ATM is a way to relax such requirements. It provides a hybrid approach to support both layer 3 software forwarding and layer 2 ATM hardware switching. A switched virtual connection may be suitable for traffic involving large data transfers and QoS provision like Guarantee Service (GS) applications. In the hybrid IP/ATM networks, the packet-level instead of the cell-level behavior is the relevant measure of performance. In this paper, we propose a so-called Frame-based Priority Scheduling (FBPS) algorithm for GS applications. FBPS focuses on the packet-level performance while transmitting based on cells. The most attractive feature of FBPS is its simplicity. We provide a systematic analysis of FBPS scheduler and derive the bound on its fairness when the flow traffic is shaped by a leaky bucket. It shows that FBPS is competitive with certain complex scheduling disciplines such as PGPS. We also verify our analytical bounds by simulation.

Research: Design and analysis of delay control for CBR services in ATM networks

Asynchronous transfer mode (ATM) networks are designed to provide end-to-end transport of user data via virtual connections with specified quality of services (QOS), which is expected to be satisfied through effective traffic control mechanisms. This paper presents an adaptive delay-jitter control (ADJC) method that guarantees the delay bounds of constant bit rate (CBR) services in ATM networks. Our proposed method is based on a node-by-node time-frame scheme. Through both mathematical derivation and simulation, we have demonstrated that ADJC features a significant improvement over existing approaches in jitter delay, which can be reduced to one time-frame from end-to-end for CBR traffic. Simulation results have shown that CBR services receive satisfactory delay performance using our approach.

An adaptive priority traffic control approach for guaranteed services in ATM networks

This paper presents an adaptive priority traffic control approach which uses a dynamic buffer allocation scheme, and an adaptive priority queuing technique for handling the delay-sensitive traffic in ATM networks. The former adjusts the buffer size based on the real traffic demand, while the latter adaptively change the priority. The simulation results show that diverse delay-sensitive traffic achieve satisfactory performance through our approach. It may fulfill the emerging needs of guaranteed services.