Nam-Seok Ko

PMIPv6 with Bicasting for IP Handover

The Proxy Mobile IPv6 (PMIPv6) was proposed to provide the Internet mobility for a host that is not equipped with any mobility software. However, the PMIPv6-based handover is still for further study. In this paper we propose a handover scheme of PMIPv6 with bicasting for IP handover. The handover performance of PMIPv6 can be enhanced with the help of bicasting in the network.

Mean starting potential fair queuing for high-speed packet networks

This paper proposes an efficient fair queuing algorithm, called mean starting potential fair queuing (MSPFQ), which has O(l) complexity for the virtual time computation while it has delay and fairness properties similar to starting-potential fair queueing (SPFQ). The key idea of MSPFQ algorithm is that it recalibrates the system virtual time to the mean value of the minimum possible virtual start times of HOL packets in each backlogged session. Through analysis and simulation, this paper shows that the proposed algorithm has good delay and fairness properties.

Emulated weighted fair queueing algorithm for high-speed packet-switched networks

In this paper we propose an efficient and simple fair queueing algorithm, called emulated weighted fair queueing (EWFQ), which has O(1) complexity for the virtual time computation while it almost perfectly emulates the delay and fairness properties of WFQ (weighted fair queueing). The key idea of EWFQ is that it calibrates the system virtual time only at the end of each packet transmission, while it updates the system virtual time for a newly arrived packet by adding elapsed real time between two calibration events to the recent system virtual time. By doing so, EWFQ has a rate-proportional property. If any fair queueing algorithm can be categorized into the rate proportional class and input is constrained by a leaky bucket, its delay is bounded and the same as that of WFQ. EWFQ can, therefore, almost perfectly emulate the delay and fairness performances of WFQ while the implementation complexity is drastically reduced to O(1). EWFQ can be implemented in a router for supporting the differential and integrated service.

An efficient anti-DDoS mechanism using flow-based forwarding technology

During the last “7.7 cyber attack”, a lot of web servers in Korea suffered from DDoS attacks even while most of the systems are believed to be guarded by various kinds of security systems. This is because it takes long to analyze and detect the attacks to filter out the traffic in most of the current anti-DDoS systems. In this paper, we propose an anti-DDoS mechanism using flow-based network forwarding technology which filters out DDoS attack traffic effectively before the other professional anti-DDoS systems can finish analysis of the attacks and react against them.

New IPv6 Transition Mechanism based on End-to-End Tunnel

Until today, there have been a lot of study for the IPv6 transition mechanism but it seems to have a lot of difficulty to expand IPv6 network due to lack of IPv6 service and unresolved interworking issues with IPv4 network. This paper proposes new IPv6 transition mechanism based on end-to-end tunneling which is expandable, easy method to deploy IPv6 services.

Bandwidth Efficient Mobile IPv6 Extension Method

With the shortage of IPv4 address space, IPv6 will be the protocol for all-IP network. In addition, as the number of mobile nodes increases, the mobility support in the protocol became the necessity in IPv6. Mobile IPv6 which is defined in RFC3375 is the basic standard to support mobility in IPv6 protocol, and there are other approaches to extend the basic mobile IPv6 architecture. The extension mechanisms are mainly focusing on addressing the hand-off latency issue, and there have been few discussions about the overhead issue in the mobility architecture. In this paper, we want to propose a bandwidth efficient mobile IPv6 extension method which requires overhead as small as possible but is still compatible with the existing architecture.

Application-level Traffic Identification using SET-based Flow in BcN

As the Internet is rapidly evolving from best-effort network to quality of service network, the important task is to provide different quality of service (QoS) guarantees to the traffic from different services in the broadband convergence network (BcN). An essential functionality for BcN would be the ability to identify applications for various purposes such as network management, QoS provisioning and access control. We developed SET-based identification method. In the SET-based method, we group all the traffic into SETs considering their end-to-end connections. In addition, we combined the method with the existing methods such as port-based and signature-based analysis in order to increase the rate of identification. We give the evaluation results in order to prove the efficiency of the hybrid method.

Performance Analysis of an IP Lookup Algorithm for High Speed Router Systems

IP lookup is an important issue in designing high speed router systems. The most important function of router is to determine the output port of an incoming IP packet according to its destination address. In this paper, we evaluate the performance of the IP lookup algorithm in IQ2200 Chipset for high speed routers in terms of the memory required for storing lookup information and the number of memory access on constructing the forwarding information and then propose a scheme to enhance the IP lookup performance.

A Novel Future-proof Software Defined Networking Architecture

This paper proposes a novel future-proof software defined networking architecture that has multiple programmable function modules that interact with flow tables. The proposed architecture can easily adapt to new protocols with variable-length header, provides new network functions locally, and let the switch serves offloaded network functions from the cloud with minimized network latency.

Design of a Multipurpose Whitebox Networking Platform

paper presents a prototype implementation of a multipurpose network platform which could be easily adapted to accommodate various types of future networks. Our design is motivated by the objective to overcome the drawbacks of existing whitebox networking devices in flexibility and performance points of view. Instead of relying on two extreme ends; ASIC or general purpose CPU, we propose a multipurpose platform which combines the moderate flexibility and rich features for fast packet processing of high-performance network processor, and almost the same flexibility as a general purpose CPU and the offloading functions such as security and deep packet inspection of multicore processor. We also show how we could use the multipurpose platform as a router with deep packet inspection feature by running an open source routing suite as control function, and provide some performance numbers for the system.