JP2003324479A - Ip simulator - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an IP simulator in which a load of a server is tested by creating a real environment wherein an error occurs in a packet without preparing a number of terminals. <P>SOLUTION: An IP simulator 3 imparts fixed or variable delay, a packet loss in a fixed cycle or a variable cycle, a burst loss quantity, and/or error of the same and the plurality of packet transmissions to a packet to be exchanged between a server 1 and a terminal 2 for each connection composed of a set of transmission source and destination addresses of an IP and transmission source and destination port numbers of a TCP or a UDP. Besides, the IP simulator has a pattern file set with the error to be imparted for each connection. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

TECHNICAL FIELD The present invention relates to an IP (Internet Pr
otocol) simulator. More specifically, the present invention relates to an IP simulator connected between a server to be tested and a client to be tested, which performs a load performance evaluation test on a server assuming access by many clients.

[0002]

2. Description of the Related Art FIG. 1 is a system configuration diagram comprising a Web server 1 and a terminal 2. Consider a case where a load performance evaluation test of the Web server 1 is desired to be simulated in the environment as shown in FIG. In particular, in an environment such as a wireless LAN, the following errors are likely to occur, so it is necessary to create such an environment.

(1) The amount of delay or the amount of packet loss differs for each connection. (2) Due to the delay, the packet arrival order is reversed. (3) Due to delay-based retransmission, multiple identical packets arrive. (4) Delay and packet loss exponentially increase according to the traffic volume.

FIG. 2 is a block diagram of a conventional simulation system for simulating the system of FIG.
According to FIG. 2, the IP simulator 3 is provided between the Web server 1 and the hub 4 that accommodates the plurality of terminals 2. The IP simulator 3 has two network interfaces (for example, Ethernet (registered trademark)), and can add errors such as delay and packet loss to packets transmitted and received between them. By adding an error to each packet in this way, an environment close to the actual Internet can be created.
It is possible to perform a load performance evaluation test on the eb server 1.

[0005]

According to the prior art,
The error given by the IP simulator 3 is IP
It was assigned for each source address and destination address of the level. This is because in the actual Internet environment, errors such as delay and packet loss occur for each transmission path, and it is sufficient to add an error at the IP level. However, in the simulation, in order to give an error at the IP level, it is necessary to prepare as many terminals 2 as the number of IPs. Therefore, in order to sufficiently perform the load performance evaluation test of the Web server 3, a large number of terminals 2 had to be prepared.

Therefore, the present invention provides an IP simulator capable of carrying out a load performance evaluation test of a server by creating an actual environment in which a packet causes an error without preparing a large number of terminals. The purpose is to do.

[0007]

The IP simulator of the present invention, for a packet to be transmitted and received, has a source address and a destination address of the IP and a TCP (Transport Control).
Protocol) or UDP (User Datagram Protocol) and a means for giving an error to each connection made up of a combination of a source port number and a destination port number.

According to another embodiment of the present invention, the error is a fixed delay amount or a variable delay amount, a packet loss at a fixed period or a packet loss at a random period, and a burst-like error at the time of packet loss. The loss of a plurality of packets and the transmission of the same and a plurality of packets for one received packet may be performed individually or in combination.

According to another embodiment of the present invention, it is also preferable to have a pattern file in which an error to be added for each connection is set.

[0010]

DETAILED DESCRIPTION OF THE INVENTION Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 3 is a block diagram of the simulation system of the present invention. According to FIG. 3, as compared with FIG. 1, it is sufficient that at least one terminal 2 is prepared, and the hub 4 is not necessary. According to the present invention, the system configuration is extremely simple.

FIG. 4 is a frame structure diagram of TCP or UDP / IP. The conventional IP simulator adds an error based on the source address and the destination address of the IP header field. On the other hand, the IP simulator of the present invention can add an error based on the transmission port number and the destination port number of the TCP header field. Further, an error can be added based on the source port number and the destination port number in the UDP header field.

Table 1 below is a comparison table of the delay applying methods. Conventionally, only one delay d [ms] can be added as a delay amount to all packets passing through the simulator. On the other hand, according to the present invention, TC
For each P or UDP connection, the delay amount is d1 ~
It becomes possible to add dn [ms].

[0014]

[Table 1]

FIG. 5 is a conceptual diagram of a protocol according to the present invention. According to FIG. 5, there is only one IP address pair between the Web server 1 and the terminal 2. However,
A plurality of connections are established by TCP or UDP. Errors such as delay and packet loss can be added to each connection. As a result, although there is only one terminal 2, it is possible to create an environment in which it is possible to communicate with a plurality of terminals.

Table 2 below is a pattern file of delay addition for each connection. According to Table 2, the following errors are added to the packet individually or in combination based on the pattern file. (1) Fixed delay amount or fluctuating delay amount (specify the maximum delay amount) (2) Packet loss for each fixed period or packet loss for each random period (3) Packet loss in (2) above Sometimes burst multiple packets are lost (4) Same and multiple packets are sent for one received packet

[0017]

[Table 2]

In Table 2, for example, in the case of the transmission line of the transmission port number a of the terminal and the transmission port number b of the Web server, a delay of 10 ms is fixedly inserted, and 60s to 1
Packet loss occurs at random intervals in 20 s, there is no burst-like loss, and the same packet is never transmitted. Also, for example, the transmission port number c of the terminal and the Web
In the case of the transmission path of the transmission port number d of the server, 10ms ~
A variable delay of 20 ms is inserted, a packet loss occurs at a fixed period of 20 s, and 10 packets are temporarily lost when the loss occurs. Send a packet.

FIG. 6 shows an IP simulator 3 according to the present invention.
2 is a protocol hierarchy configuration diagram of FIG. 6, a network interface layer 31 including a physical layer and a data link layer, an internet layer 32 for controlling IP or IPv6, a transport layer 33 for controlling TCP or UDP, and a pattern file 341 are provided. And an error giving unit 34 that gives an error.

The source address and the destination address of the packet received by the network interface layer 31 are detected by the Internet layer 32. Next, in the transport layer 33, the source port number and the destination port number of the packet are detected. As a result, a packet is detected for each connection. Next, the error adding unit 34 adds a delay or packet loss error for each connection based on the pattern file 341. The delayed packet is transmitted to the server 1 via the transport layer 33, the internet layer 32, and the network interface layer 31. On the other hand, the packet considered as packet loss is discarded as it is.

In the embodiment of the IP simulator according to the present invention described above, those skilled in the art can easily make various changes, modifications and omissions of the technical idea and scope of the present invention. The above description is merely an example and is not intended to be limiting. The invention is limited only by the claims and equivalents thereof.

[0022]

As described above in detail, by using the IP simulator of the present invention, it is possible to create an actual environment in which a packet causes an error without preparing a large number of terminals, and It becomes possible to perform a load performance evaluation test.

[Brief description of drawings]

FIG. 1 is a system configuration diagram including a server and a terminal.

FIG. 2 is a configuration diagram of a conventional simulation system.

FIG. 3 is a configuration diagram of a simulation system of the present invention.

FIG. 4 is a frame configuration diagram of TCP or UDP / IP.

FIG. 5 is a conceptual diagram of a protocol according to the present invention.

FIG. 6 is a protocol hierarchy diagram of an IP simulator according to the present invention.

[Explanation of symbols]

1 Web server 2 terminals 3 IP simulator 31 Network Interface Layer 32 Internet layer 33 Transport layer 34 Error giving section 341 pattern file 4 hubs

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Hideyuki Shinonaga             2-15-1 Ohara, Kamifukuoka City, Saitama Stock             Company CAD Research Institute F term (reference) 5B089 GA11 GA21 GB02 HA10 JB19                 5K030 GA12 HB12 HB17 JA10 KX30                       LD20 MC03                 5K033 AA05 CB06 CB08 DB20

Claims (3)

[Claims] 1. An IP simulator connected between a server to be tested and a client to be tested, for a packet to be transmitted / received, an IP source address and a destination address, and TCP or UDP. I has a means for giving an error to each connection consisting of a combination of a source port number and a destination port number of
P simulator. 2. The error includes a fixed delay amount or a variable delay amount, a packet loss for every fixed period or a packet loss for every random period, a burst-like loss of a plurality of packets at the time of packet loss, and reception. 2. The IP simulator according to claim 1, wherein the IP simulator is a single packet or a combination of transmission of the same and a plurality of packets. 3. The IP simulator according to claim 1, further comprising a pattern file in which the error to be added for each connection is set.