When using per-vu-iteration for dispatching concurrent requests, for example, having ‘vus’ set to the number of concurrent requests and iteration to be executed per vu set as ‘1’, it was observed that the requests are not being dispatched at the same time. For example, when the number of vus was set to 500, with each vu executing 1 iteration each, looking at the requests per second, we are seeing the requests dispatched in different one second time block. Is there a reason why the requests are not being sent at the same time? Is there other way we can use k6 to push concurrent requests? i.e. we want to see how the api will behave if we push 500 requests at the same time
Hi @angmenghua, welcome to the community forum and sorry for the slow reply 
Can you provide the sample script or at least the options block as more info.
Also it will be nice to have CPU(and memory usage) on the load generator (and likely the SUT). If either of those is maxed out that might explain why the 500 requests isn’t hit.