Rueidis Compat Client vs Go-Redis latency

[t-kruth]

Currently we're looking to leverage rueidis in place of the go-redis package. We leveraged it in a service and noticed that the roundtrip latency spiked along with our processing of events.

Currently running version 7.1 redis in AWS Elasticache in clustermode

we're using mostly default config options with the exception of
BlockingPoolSize = 100
Dialer.Timeout = 1s
SenToReplicas = cmd.IsReadOnly()
and leveraging TLS

the client is created,
we add a hook for timing
and then leverage the rueidiscompt.NewAdapter so we can leverage the already existing code for redis operations

[rueian]

we add a hook for timing

Do you mean rueidishook?

so we can leverage the already existing code for redis operations

What redis operations are you using?

We leveraged it in a service and noticed that the roundtrip latency spiked along with our processing of events.

Would it be possible to add more CPU cores to your application and see if those spikes go down?

[t-kruth]

Do you mean rueidishook?

Yup, we're using a rueidishook.WithHook to add the timing func

What redis operations are you using?

Current Redis Operations

SetNX
ZAdd
Expire
ZRem
ZRangeByScore
ZRemRangeByScore
Del

Would it be possible to add more CPU cores to your application and see if those spikes go down?

Yea we should be able to give some more resources to the service. I'll give that a try and update this thread on what we see

[rueian]

One possible explanation for those spikes might be large responses from the ZRangeByScore causing delays in the same connection pipeline. You could also try a higher PipelineMultiplex which opens multiple connections (2^n) to the same redis node. By doing that those blocking delays should be relieved as long as you are not exhausting your CPUs.

client, err := rueidis.NewClient(rueidis.ClientOption{
	InitAddress:       []string{"aws_configuration_endpoint"},
	PipelineMultiplex: 1, // open 2 connections to every node.
})

Note that

1. PipelineMultiplex: 1 opens 2 connections to every node.
2. PipelineMultiplex: 2 opens 4 connections to every node.

PipelineMultiplex>=3 is not recommended.

[t-kruth]

ok, i'll add that to the things to try. I thought we saw somewhere documented that if redis is in cluster mode, pipelinemultiplex is ignored?

[rueian]

You are right, but we removed the ignorance of PipelineMultiplex for cluster clients since https://github.com/redis/rueidis/releases/tag/v1.0.24.