Load balancing vusers without a load balancer

Recently I ran a test at a company which had a performance test environment with multiple web servers, but no load balancer. To spread my virtual users evenly across the web servers, I made a simple modification to my script.

First, I created a file-based parameter containing my web server names.
VuGen Parameter List showing parameter-based load balancing

Next, I parameterised the server name part of the URLs in my web_url() and web_submit_data() function calls.

1
2
3
4
5
6
7
8
9

web_url("Login", 
    "URL=http://{ServerName}/login.do", 
    "TargetFrame=", 
    "Resource=0", 
    "RecContentType=text/html", 
    "Referer=", 
    "Snapshot=t1.inf", 
    "Mode=HTML", 
    LAST);

So, when the script ran, the virtual user would would choose a web server at random on every iteration.

…and that was it. I didn’t say it was going to be hard. :)

Of course, my Test Summary Report was very clear that one of the limitations of the Performance Test cycle had been that there was no load balancer in the test environment, so the stability and performance characteristics of this component in the Production environment could not be predicted.

5 comments

Rajesh sahu
June 19, 2012 Reply

Suppose I have 1 web server and 2 Application server then also in this case load balancer concept will come into picture.

Because URL hits to web server then it goes to application server and then database server. So I think we can not divide the load on Application servers.

Regards,
Rajesh Sahu
rajeshbscity6@gmail.com

Stuart Moncrieff
June 25, 2012 Reply

Hi Rajesh,

You are correct for IP-based load balancing. Sometimes load is balanced to the app servers based on a cookie value you can manipulate.

Sometimes I bypass the web server tier and hit the app servers directly. In your case it would be best to set up multiple IP addresses and use IP Spoofing.

Cheers,
Stuart.

Smruti
August 26, 2016 Reply

Will it work if i run the scenario through Peformance Center.

Regards,
Smruti

vdaburon
April 17, 2009 Reply

A alternative solution uses the vuser id modulo the number of hosts.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

	static int NB_HOST = 3; // 3 differents hosts
 
	/* load balancing computes with vuser id modulo number of hosts */
	int id, scid;
	char *vuser_group;
	int valIdMod;
	static int isInitRand = 0;
 
	if (isInitRand == 0) {
		//srand is called before rand 
		srand(time(NULL)); 
		isInitRand = 1;
	}
 
	lr_whoami(&id, &vuser_group, &scid);
	lr_message( "Group: %s, vuser id: %d, scenario id %d",  vuser_group, id, scid);
 
	/*************** BEGIN COMMENT ***********/
	// id = 5555; // force the id to a specifique host
	/*************** END commenter ***********/
 
	valIdMod = id % NB_HOST;
 
	if (valIdMod == 0) {
		lr_save_string( "http://","P_PROTOCOLE" );
		lr_save_string("serverA","P_SERVEUR" ); 
		lr_save_string( "7013","P_PORT" );
	}
 
	if (valIdMod == 1) {
		lr_save_string( "http://","P_PROTOCOLE" );
		lr_save_string("serverB","P_SERVEUR" ); 
		lr_save_string( "8080","P_PORT" );
	}
 
	if (valIdMod == 2) {
		lr_save_string( "http://","P_PROTOCOLE" );
		lr_save_string("serverC","P_SERVEUR" ); 
		lr_save_string( "7015","P_PORT" );
	}
 
	// force to a specific host
	if (id == 5555) {
		lr_save_string( "http://","P_PROTOCOLE" );
		lr_save_string("serverB","P_SERVEUR" ); 
		lr_save_string( "8080","P_PORT" );
 
	}
	lr_log_message("P_SERVEUR = %s", lr_eval_string("{P_SERVEUR}"));
	lr_log_message("P_PORT = %s", lr_eval_string("{P_PORT}"));
 
	web_url("Login", 
		"URL={P_PROTOCOLE}://{P_SERVEUR}:{P_PORT}/login.do", 
		"TargetFrame=", 
		"Resource=0", 
		"RecContentType=text/html", 
		"Referer=", 
		"Snapshot=t1.inf", 
		"Mode=HTML", 
		LAST);

Leave a Reply