Istio performance and scalability

Hello everyone

In relation to the official documentation about performance and scalability. Could someone tell me what Istio configuration they used to get these results?

I am testing in a cluster with 1250 pods and injecting a new service with sidecar and the statistics show me that the sidecar consumes in average 160 MB of memory.

I must mention that I installed Istio with this command:

helm template install / kubernetes / helm / istio --name istio --namespace istio-system
–values install / kubernetes / helm / istio / values-istio-demo.yaml --set global.proxy.concurrency = 1 | kubectl apply -f -

  • Istio install option: Demo
  • global.proxy.concurrency = 1, this I read is to reduce memory consumption.

Well I repeat the question: Could someone tell me what configuration of Istio they used to obtain these results?
50 MB per sidecar in a large namespace, its a dream for me.