Active mutations total

What I want to do

  1. How to adjust startup parameters to improve throughput?
  2. And why does the dgraph_pending_proposals_total indicator show that one node reaches the critical value of 256, but the other nodes are still 0?

What I did

I use jmeter to perform pressure test on dgraph-alpha-public/mutate.Under 1000 concurrency, the throughput of 20 triples on my side is 1500 for each mutation; the throughput is 4000 for each mutation of 6 triples; the throughput is 4000 for each mutation of two triples;

Dgraph metadata

dgraph version

Dgraph version : v20.11.2
Dgraph codename : tchalla-2
Commit timestamp : 2021-02-23 13:07:17 +0530
Branch : HEAD
Go version : go1.15.5
jemalloc enabled : true

  1. what are the flags you use currently?
  2. Pending proposals are basically related to raft. It could mean a few things to be honest. Your CPU saturation could be high, or there could be network issues between your nodes. Can you check the /health endpoint to see if everything is healthy?

I currently set --pending_proposals to 512, but there is no obvious difference compared with 256.

I see that the health status is normal from the /health endpoint, and it is also normal to view the dgraph_alpha_health_status indicator. I am trying to get the usage indicator of each pod cpu.