Is deploying multiple instances of Tyk OSS in AWS as simple as increasing desired count?

[timothy_park]

I just had a quick question about deploying multiple instances of Tyk OSS in AWS ECS. We want to scale up the number of instances that are handling traffic to ensure that we don’t have a bottleneck with our largest prod environment. Can we simply increase the desired count and will key permissions and rate limits all be controlled through their connection with redis? And if so, could we use this feature to auto-scale up resources when CPU utilization becomes too high?

[Olu]

I am not sure how the gateway is deployed in AWS ECS but deploying multiple instances of Tyk is one thing, syncing them is another. There are a couple of similar discussions about this on the forums

• Replication Between TYK Servers
• How to syn 2 tyk open source gateway?
• Tyk OSS in Kubernetes with more ReplicaCount
• How to scale Tyk OpenSource Gateway?
• Tyk Operator with multiple Gateway replicas

@timothy_park

Can we simply increase the desired count and will key permissions and rate limits all be controlled through their connection with redis?

I guess it’s possible if reds rate limiter is used

@timothy_park

And if so, could we use this feature to auto-scale up resources when CPU utilization becomes too high?

What feature would this be and how does access control correlate with CPU utilization?