@andytanmc, okay, @cusello gave you a simple and useful answer, but the more important context question is, what are you trying to achieve by simulating concurrent searches? That strategy will spit out 100 searches, and your system will do whatever it is set up to do when one person submits 100 searches. If that tests your system as you want, then fine. On the other hand, it may be more accurate to, for example, invent a single search that typically takes 5 minutes to complete, create 25 simple variations of that search that won't interact with each other, and set them each to run on a rolling scheduled basis every 5 minutes under 5 different IDs. (So ID 1 submits first search at minutes 1, 6, 11 etc and second search at minutes 2, 7, 12 etc), and there are always about 25 searches going. That gives you a "floor" of roughly 25 searches. When that is working, pop up a dashboard with a different user running 25 searches, and see what happens. ... View more