• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.44 no.9
• /
• pp.38-45
• /
• 2007

With the emergence of new applications, packet classification is essential for supporting advanced internet applications, such as network security and QoS provisioning. As the packet classification on multiple-fields is a difficult and time consuming problem, internet routers need to classify incoming packet quickly into flows. In this paper, we present multi-phase packet classification architecture using an internal buffer for fast packet processing. Using internal buffer between address pair searching phase and remained fields searching phases, we can hide latency from the characteristic that search times of source and destination header fields are different. Moreover we guarantee the improvement by using single entry caching. The proposed architecture is easy to apply to different needs owing to its simplicity and generality.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.3B
• /
• pp.339-348
• /
• 2004

Packet classification is an important factor to support various services such as QoS guarantee and VPN for users in Internet. Packet classification is a searching process for best matching rule on rule tables by employing multi-field such as source address, protocol, and port number as well as destination address in If header. In this paper, we propose hardware based packet classification algorithm by employing tree bitmap of multi-bit trio. We divided prefixes of searching fields and rule into multi-bit stride, and perform a rule searching with multi-bit of fixed size. The proposed scheme can reduce the access times taking for rule search by employing indexing key in a fixed size of upper bits of rule prefixes. We also employ a marker prefixes in order to remove backtracking during searching a rule. In this paper, we generate two dimensional random rule set of source address and destination address using routing tables provided by IPMA Project, and compare its memory usages and performance.

• Journal of Communications and Networks
• /
• v.3 no.4
• /
• pp.324-333
• /
• 2001

The Differentiated Services (Diffserv) framework has been proposed by the IETF as a simple service structure that can provide different Quality of Service (QoS) to different classes of packets in IP networks. IP packets are classified into one of a limited number of service classes, and are marked in the packet header for easy classification and differentiated treatments when transferred within a Diffserv domain. The Diffserv framework defines simple and efficient QoS differentiation mechanisms for the Internet. However, the original Diffserv concept does not provide a complete QoS management framework. Since traffic flows in IP networks are unidirectional from one network point to the other and routing paths and traffic demand get dynamically altered, it is important to monitor end-to-end traffic status, as well as traffic status in a single node. This paper suggests a distributed QoS monitoring method that collects the statistical data of each service class in every Diffserv router and calculates edge-to-edge QoS of the aggregated IP flows by combining routing topology and traffic status. A format modeling of edge-to-edge Diffserv flows and algorithms for aggregating edge-to-edge QoS is presented. Also an SNMP-based QoS management prototype system for Diffserv networks is presented, which validates our QoS management framework and demonstrates useful service management functionality.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2003.11b
• /
• pp.1053-1056
• /
• 2003

라우터에서 패킷을 특정 "플로우"로 분류하는 과정을 패킷 분류라고 한다. 어떤 플로우에 소한 모든 패킷들은 미리 정의된 규칙을 따르며. 라우터는 비슷한 처리과정을 제공한다. 예를 들면, 같은 줄발지 IP주소와 도착지 IP주소를 가진 패킷들은 하나의 플로우로 분류된다. 이러한 패킷 분류는 최선형(best-effort) 서비스가 아닌 품질보장(QoS), VPN, 방화벽 서비스 등과 같이 플로우와 플로우를 구별하여야하는 서비스 제공에 필요하다. 일반적으로 다수의 필터에 대한 패킷 분류는 매우 어려운 문제로 하드웨어 및 소프트웨어 기반의 다양한 알고리즘들이 제시되었다. 본 논문은 트라이를 기반으로 작은 메모리 공간을 사용하면서 빠른 패킷 분류를 탄 수 있는 기법을 제시한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.11C
• /
• pp.1164-1170
• /
• 2002

In the ACE2000 MPLS system, MPLS Interface Module(MIM) is composed of an ATM Interface Module and a HFMA performing a packet forwarding. In the MIM, the HFMA RSAR receive cells from the Physical layer and reassemble the cells. And the IP Lookup controller perform a packet forwarding after packet classification. Forwarded packet is segmented into cells in the HFMA TSAR and transfer to the ALMA for the transmission to an ATM cell switch. When the MIM make use of an ATM Interface Module, it directly connect the ALMA with a PHY layer using the UTOPIA Level2 interface. Then, an ALMA performs Master Mode. Also, the HFMA TSAR performs the Master Mode in the MIM. Therefore, the UTOPIA-L2 Controller of the Slave Mode require for interfacing between an ALMA and a HFHA TSAR. In this paper, we implement the architecture and cell control mechanism for the UTOPIA-L2 Controller supporting Multi-ports.

• Journal of KIISE:Information Networking
• /
• v.28 no.2
• /
• pp.251-261
• /
• 2001

The traffic on the Internet has been growing exponentially for some time. This growth is beginning to stress the current-day routers. However, switching technology offers much higher performance. So the label switching network which combines IP routing with switching technology, is emerged. EspeciaJJy in the data driven label switching, flow classification and cache table management are needed. Flow classification is to classify packets into switching and non-switching packets, and cache table management is to maintain the cache table which contains information for flow classification and label switching. However, the cache table management affects the performance of label switching network considerably as well as flowclassification because the bigger cache table makes more packet switched and maintains setup cost lower, but cache is restricted by local router resources. For that reason, there is need to study the cache replacement scheme for the efficient cache table management with the Internet traffic characterized by user. So in this paper, we propose several cache replacement schemes for label switching network. First, without the limitation at switching capacity in the router. we introduce FIFO(First In First Out). LFC(Least Flow Count), LRU(Least Recently Used! scheme and propose priority LRU, weighted priority LRU scheme. Second, with the limitation at switching capacity in the router, we introduce LFC-LFC, LFC-LRU, LRU-LFC, LRU-LRU scheme and propose LRU-weighted LRU scheme. Without limitation, weighted priority LRU scheme and with limitation, LRU-weighted LRU scheme showed best performance in this paper.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.8
• /
• pp.327-334
• /
• 2013

Today's network traffic has become extremely complex and diverse since the speed of network became faster and a variety of application services appear. Moreover, many applications appear and disappear fast and continuously. However, the current traffic classification system does not give much attention to this dynamic change of applications. In this paper, we propose an application awareness system in order to solve this problem. The application awareness system can provide the information, such as the usage trend of conventional applications and the emergence of new applications by recognizing the application name in a rapidly changing network environment. In order to recognize the application name, the Host field of HTTP protocol has been utilized. The proposed mechanism consists of two steps. First, the system generates the candidates of application name by extracting the domain name from the Host field in HTTP packet. Second, the administrator confirms the name afterward. The validity of the proposed system has been proved through the experiments in campus network.