KR100543446B1 - Performance management method for traffic measuring of label switching path - Google Patents

Abstract

The present invention measures traffic of each label switching path in multiprotocol label switching (MPLS) and measures traffic of a label switching path that can efficiently perform performance management. It relates to a performance management method for. The performance management method for measuring traffic of the label switching path of the present invention includes an input and an output router that receives traffic from a user computer and connects the label switching path of a multi-protocol label switching network to and from the input and output routers. Routers consisting of core routers that relay traffic, located in different domains, are connected to the Fast Ethernet ports of the input and output routers via hubs, and one of them has a network management system function. Configuring a multi-protocol label switching network with a plurality of user computers loaded, creating a label switching path between the input router and the output router, and a separate label switching path or the Fast Ethernet port flowing into the label switching path. For monitoring Measuring traffic of a label switching path, creating an access object using a command interface module of a generic connection adapter (GIA) to lead the measured traffic, and using the access object to measure the measured traffic Reading and transmitting at least one or more information of a label switching path identifier, IP information of both routers for which the label switching path is set, allocation bandwidth of the label switching path, and delay time information to the network management system. And monitoring the transmission information in the network management system.

Traffic Engineering, MPLS

Description

Performance management method for traffic measuring of label switching path

1 is a diagram showing a multi-protocol label switching (MPLS) network configuration according to the present invention

2 is a view for explaining a performance management method for measuring traffic in a multi-protocol label switching network according to the present invention.

* Description of the symbols for the main parts of the drawings *

10, 20: hub 30, 40, 50: router

60,70,80,90,100: User computer 110: MPLS network

120: general connection adapter 200: network management system

210: IP layer 220: data link layer

230: physical layer 240: MIT handler

250: MIT 260: GUA

270: Java GUI 280: Web Reporter

290: database

The present invention measures traffic of each label switching path in multiprotocol label switching (MPLS) and measures traffic of a label switching path that can efficiently perform performance management. It relates to a performance management method for.

Multiprotocol Label Switching (MPLS) network technology developed to improve high-speed packet transmission and switching performance using labels is a label switching that allows bandwidth reservation between Label Switching Routers (LSRs). It provides a more flexible service by providing a path (Label Switching Path (abbreviated as LSP)) and traffic engineering.

This flexible traffic engineering capability of MPLS enables the provision of multimedia services that require real-time characteristics and quality of service (QoS) guarantees.

In addition, in order to manage MPLS network, administrator must move to each node's switch and input command through Command Line Interface (CLI). In such a case, know all difficult commands It was possible to deal with.

However, as the number of switches and switching equipment increases dramatically, there is a limit to tasks requiring human hand, and the system is maintained by inefficient repetition. Therefore, it is necessary to control through the management of an efficient network management system (NMS) that can replace people.

The present invention has been made to solve the problems of the prior art, and automates the part of setting and releasing the LSP in the MPLS network by the NMS, and for each LSP, a traffic parameter and a quality of service parameter (QoS parameter). The purpose is to provide a performance management method for traffic measurement of a label switching path that can control the order through priority for each LSP.

A feature of the present invention for achieving the above object is an input and output router for receiving traffic from a user computer and connecting to a label switching path of a multiprotocol label switching network, and a core for relaying traffic between the input and output routers. Routers configured as routers, located in different domains, connected to the Fast Ethernet ports of the input and output routers via hubs, and one of them is loaded with a network management system function. Configuring a multi-protocol label switching network with a computer, creating a label switching path between the input router and an output router, and monitoring the individual label switching path or the Fast Ethernet port flowing into the label switching path. Switching path Measuring an access point; generating an access object by using a command interface module of a generic connection adapter (GIA) to lead the measured traffic; and reading the measured traffic by using the access object; Transmitting, by the GIA, at least one or more of a path identifier, IP information of both routers for which the label switching path is set, allocation bandwidth and delay time information of the label switching path, to the network management system; And monitoring the transmission information in a management system.

Preferably, the monitoring of the transmission information in the network management system is performed in the internet protocol layer, data link layer, and physical layer in the network management system.

Other objects, features and advantages of the present invention will become apparent from the following detailed description of embodiments taken in conjunction with the accompanying drawings.

Hereinafter, a performance management method for measuring traffic of a label switching path of the present invention will be described with reference to the accompanying drawings.

1 is a diagram showing a multi-protocol label switching (MPLS) network configuration according to the present invention.

MPLS network 110 configuration according to the present invention, as shown in Figure 1, is located in different domains (Domain), and through the hub (10, 20) Fast Ethernet (each of the routers 30, 40, 50 ( Fast Ethernet) is composed of a plurality of first to fifth user computers (hereinafter, abbreviated as users) (60, 70, 80, 90, 100) connected to the port, of which the first user computer 60 It loads the network management system function, where the network management system is programmed to be accessible to each router (30, 40, 50).

Here, the core router 30 plays a role of relaying traffic between the PE1 (Provide Edge) router 40 and the PE2 router 50, and the PE1 (Provide Edge) router 40 and the PE2 router 50 Receives traffic from users 60, 70, 80, 90, and 100 as input and output (or output and input) routers, respectively, and maps them to MPLS LSPs (paths), and transmits them through the core router 30. Extract the user traffic from the output router for MPLS traffic.

Define a class-map to specify the baud rate, packet loss rate, and differentiated services for each user (command interface (CLI) command-class map, through GIA programming). Definition of a policy map, etc.) and class information is input to classify packets coming into a fast Ethernet port of the router 40 or 50 into a specific class. A policy-map is defined to apply each class-map.

That is, in order to assign a differentiated service (DiffServ) to a class classified in an actual MPLS network 110, a policy-map corresponding to a port of each router must be specified.

The basic setting to be used in the actual network configuration is to classify the service class for each user into classes (EF, AF4, AF1) as follows.

Here, the service level may be classified as follows according to priority.

NCT: Network Control Traffic

EF: Expedited Forwarding

AF1-4: Assured Forwarding

BE: Best Effort

This is shown in Table 1

policy-map SETDSCP  class EF set ip dscp 46  class AF1 set ip dscp 10  class AF2 set ip dscp 18 class AF3 set ip dscp 26  class AF4 set ip dscp 34 class BE  set ip dscp 0

In order to manage the actual MPLS network 110 through the NMS, an efficient NMS system is required.

By using an automated command line interface through the NMS system, the MPLS network 110 may have automatic bandwidth allocation, automatic allocation of LSPs, and an efficient reporting system.

2 is a view for explaining a performance management method for measuring traffic in a multi-protocol label switching network according to the present invention.

In order to effectively manage the lower MPLS network 110, the NMS 200 is divided into hierarchical, that is, the IP layer 210, the data link layer 220 and the physical layer 230, so that the overall details of the detailed equipment and the network can be obtained. To manage.

First, the IP connection performer (IPCP) in the IP layer 210 is a connection management module of NMS in charge of connection management of each node in the IP layer 210 belonging to the third layer of the OSI (Open Systems Interconnection) 7 layer, IP Topological configurator (IPTC) is a configuration management module of NMS that manages the configuration of each node and link in the IP layer 210 belonging to the third layer of the OSI (Open Systems Interconnection) 7 layer, and IP Performance Management (IPPM) Performance management of NMS that is responsible for monitoring, analysis, control, and tuning of overall traffic in IP layer 210 belonging to three layers of OSI (Open Systems Interconnection). IP Fault Management (IPFM) is an NMS fault management module that is responsible for determining failure and recovery of nodes and links in the IP layer 210 belonging to the third layer of the OSI (Open Systems Interconnection) 7 layer.

In the data link layer 220, a data link connection performer (DLCP) is an NMS connection management module that manages connection of each node in the data link layer 220 belonging to the second layer of the OSI 7 layer. , DLTC (DataLink Topological Configurator) is a configuration management module of NMS that manages the configuration of each node and link in the data link layer 220 belonging to the second layer of OSI 7 layer, DLPM (Data Link Performance Management) is OSI 7 NMS's performance management module responsible for monitoring, analyzing, controlling, and tuning the overall traffic in the data link layer 220 belonging to the second layer of the layer, and Datalink Fault Management) is an NMS fault management module that is responsible for determining faults and recovering nodes and links in the data link layer 220 belonging to the second layer of the OSI 7 layer.

In addition, the Physical Topological Configurator (PhyTC) in the Physical Layer 230 is a configuration management module of the NMS that manages the configuration of each node and link in the Physical Layer 230 belonging to the first layer of the OSI 7 layer, PhyPM (Physical Performance Management) is responsible for monitoring, analyzing, controlling, and tuning the overall traffic in the physical layer 230 belonging to the first layer of the OSI 7 layer. NMS performance management module, PhyFM (Physical Fault Management) is the NMS fault management responsible for node and link failure determination, recovery, etc. in the physical layer 230 belonging to the first layer of OSI 7 layer Module.

In addition, the web reporter 280 of the user interface 270 stores daily, weekly, monthly and yearly usage data (disturbance, performance, usage status, etc.) of the NMS in the database 290 to store the data. Allows the user / administrator to know remotely.

The Management Information Tree (MIT) handler 240 is a service program that makes the interface easier to access and use the MIT service containing the CORBA object address, and the COBRA naming service (MIT) 250 is a distributed CORBA object. It is a database in the form of a collection of structures that are responsible for their addresses.

The generic interface adapter (GIA) 120 periodically uploads traffic monitoring information and resource monitoring information of each host to the NMS 200, which is generated by the PC1 to PC6 below. In step 200, the command requested to the MPLS network 110 is transferred to a Real Monitoring Agent (RMA) interface, which is an internal module, to perform a necessary traffic monitoring function.

The MPLS Performance Management function having such a configuration will be described.

First, there is direct access to the MPLS routers that make up the underlying network.

At this time, the access object is created using the command interface (CLI) module of the GIA 120.

In addition, there is a function of reading MPLS router information of the MPLS network 110.

For example, getLSPInfo () is a method related to LSP configuration information. When the GetLSP method is called, LSPId, router IP information at both ends of LSP is set, LSP allocation bandwidth, delay time, etc. are stored in a structure variable called returnLSPInfo. Return it.

And LSP generation.

Where LSP is the path (LSP) between the ingress router and the egress router.

For example, CreateLSPDynamic dynamically creates an LSP, CreateLSPExplicit creates an LSP with an explicit specification, and CreateLSPQTH creates an LSP with a Routing Algorithm (QTHR).

There is also a delete function of the LSP.

For example, releaseLSP () is a connection establishment or release, and m_lpCISCO-> NoInterfaceTunnel (tmp_LSPId) releases the request LSPid through the method.

It also provides monitoring of traffic flowing into the LSP.

For example, monitoringLSP () is the monitoring of individual LSPs, and monitoringPort () measures the traffic across the LSP by monitoring the ports.

Therefore, according to the present invention, the Differentiated Service Code (DSCP) mapped in the DiffServ Domain (router under the general connection adapter (GIA) 120 of FIG. 2) of the multi-protocol label switching (MPLS) network 110 is used. Point values and label information are read to allocate resources for providing a corresponding service, and data is transferred to an MPLS edge router (DS egress).

The egress router removes the label assigned to the packet and then forwards the packet to the destination user.

The MPLS network 110 is configured through such a series of processes, and the traffic is allocated as agreed with the SLA (Service Label Agreement) for each LSP in the network providing the differentiated service (DiffServ). Each module in the IP layer 210, the data link layer 220, and the physical layer 230 of the NMS 200 measures the performance of the differentiated service (DiffServ) well by providing the ranking.

In this case, using an automated command line interface (CLI) through the NMS 200, the MPLS network 110 may have automatic bandwidth allocation, automatic allocation of LSPs, and an effective reporting scheme.

In order to effectively manage the lower MPLS network 110, the NMS system is divided into hierarchical layers (IP layer 210, data link layer 220, and physical layer 230) as described above. It is possible to manage the overall details of the, traffic monitoring information and resource monitoring information of each host that occurs at the bottom periodically sends to the management system (NMS) (200), NMS (200) ) Transmits the command requested to the MPLS network 110 to the NMS internal modules to perform the necessary traffic monitoring function.

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the spirit of the present invention.

Therefore, the technical scope of the present invention should not be limited to the contents described in the embodiments, but should be defined by the claims.

As described above, the present invention shows whether the traffic is properly allocated to each label switching path in the network providing the differentiated service (DiffServ) and whether the characteristics of the differentiated service (DiffServ) are well represented through prioritization. It is possible to manage the overall details of the detailed equipment and network in the IP layer, data link layer and physical layer, and periodically manages the resource monitoring information and resource monitoring information of each host. If you send it to the network, the management system has the effect of performing the necessary traffic monitoring functions.

Claims (2)

Input and output routers that receive the traffic from the user's computer and connect them to the label switching path of the multi-protocol label switching network, routers composed of core routers that relay traffic between the input and output routers, and different domains. And a multi-protocol label switching network configured with a plurality of user computers loaded with a network management system function, which is connected to a Fast Ethernet port of the input and output router via a hub. ; Creating a label switching path between the input router and the output router; Measuring the traffic of the label switching path by monitoring the individual label switching path or the Fast Ethernet port flowing into the label switching path; Creating an access object using a command interface module of a generic connection adapter (GIA) to lead the measured traffic; The measured traffic is read using the access object to transmit at least one or more of a label switching path identifier, IP information of both routers for which the label switching path is set, allocation bandwidth of the label switching path, and delay time information. Transmitting by the GIA) to the network management system; In the network management system, an Internet protocol layer including IP Connection Performer (IPCP), IP Topological configurator (IPTC), IP Performance Management (IPPM), and IP Fault Management (IPFM), Data Link Connection Performer (DLCP), and DLTC ( Data Link Layer including DataLink Topological Configurator (DLPM), Data Link Performance Management (DLPM), Datalink Fault Management (DLFM) Monitoring the transmission information hierarchically Performance management method for traffic measurement of the label switching path comprising a. delete

Images