How to config to serve large scale request

[kingstonduy]

Hi @tsurdilo,

I find your repository very useful. However when i use docker to run the your set up temporal server

docker network create temporal-network
docker compose -f compose-postgres.yml -f compose-services.yml up --detach

When i start 1000 workflows, each workflow has 8 activites. the average time for each request is 8 seconds (is it normal? To me it is insanely slow).
This is the code i set up the worker in go:

	workers := 100
	var forever chan int
	for i := 0; i < workers; i++ {
		go func() {
			// This worker hosts both Workflow and Activity functions
			w := worker.New(*bootstrap.GetTemporalClient(mw.cfg), mw.cfg.Temporal.TaskQueue, worker.Options{
				MaxConcurrentActivityTaskPollers: 10,
				MaxConcurrentWorkflowTaskPollers: 10,
				Identity:                         uuid.New(),
			})

			w.RegisterWorkflowWithOptions(usecase.MoneyTransferWorkflow, workflow.RegisterOptions{
				Name: "MoneyTransferService",
			})

			w.RegisterActivity(activity1)
			w.RegisterActivity(activity2)
                        w.RegisterActivity(activity3)
                        w.RegisterActivity(activity4)
                        ...
                        w.RegisterActivity(activity7)
                        w.RegisterActivity(activity8)
			// Start listening to the Task Queue
			err := w.Start()
			if err != nil {
				log.Fatalln("unable to start Worker", err)
			}
		}()
	}
	<-forever
}

So my question is: Did i config the worker in a wrong way or i just need to scale the temporal if it is can you specify how to scale it properly.

Thank you for reading.

[tsurdilo]

Hi, sorry for late reply. yeah we should try to figure out whats going on as in do we need to scale cluster, or is db just slow or is proxy (envoy) the culprit.
Would start looking at persistence latencies and server latencies. Can do a test on this using your code and start checking. ty for reporting this
Yes this is just a starting point for things not meant as a scale test out of box solution