amazon web services - Issues while testing on AWS Auto Scaling, ELB, Cloud Watch -

i created web app in jsp. 1 of web app url return unique id.

here url. www.biomobilestrokelab.com/gatekeeper/newuserid.jsp

it works fine in web browser. want test scalablity on web app. use apache jmeter purpose. when hit 1000 requests per second

• sometimes requests return response successfully
• sometimes receive http 504 gateway timeout code
• somtimes receive http 503 service unavailable back-end server @ capacity.

i using aws autoscalig

1. minimum instances = 2
2. maximum instance = 12
3. health check grace period = 300 sec
4. default cool down = 60 sec

for elb follwing options configured.

1. time out = 60 sec
2. interval = 200 sec
3. unhealthy threshold = 2
4. healthy threshold = 10

and apply following cloud watch matrices auto scaling.

1. cpu utilization: add 1 instance when greater 10% , remove 1 instance when less 3% , applied average statistics period of 1 minute.

2. request count: adds 1 instance when sum greater 1000 , decrease 1 instance when sum less equal 1000.

kindly guide me how can resolve issue can hit 1000 or more 1000 requests per second.

i suggest either reduce number of requests per second fixed number of instances (to let's two) or increase number of instances until there no error , thousand requests per second handled successfully. based on this, can set min in auto scaling group , adjust cloudwatch alarm.