Elasticache Single node - Read endpoint

Question

Good morning,  
  
Can anyone help. I have noticed some slow logs in our Redis Elasticache DB which are associated with some rather slow hgetall requests coming from Python. I had seen the Redis read endpoint so thought I would point the python making these queries to that end point instead of our primary end point however I still see the requests stacking in the slow log. Does this mean the read end point is actually just pointing at the same instance?  
  
Clustering : off  
Replication : off  
  
Single node.  
  
Thanks  
  
Tom

Accepted Answer

Hey Tom,  
  
A reader endpoint will split incoming connections to the endpoint between all read replicas in a Redis cluster. Reader endpoints keep up with cluster changes in real-time as replicas are added or removed. You can place your Redis cluster’s multiple read replicas in different AWS Availability Zones (AZ) to ensure high availability of reader endpoints.  
  
Without replication, the Reader endpoint will be pointed to the only available node on the cluster, to take advantage of read-scalability and to create fail-over target, we would recommend setting up a replica to balance the load off of the primary node for these read requests. Otherwise, perhaps running HSCAN to iterate through the Hash instead of HGETALL might keep the server from being blocked.  
  
Best,  
Zach

Elasticache Single node - Read endpoint

Relevant content