Hi,
We're working on an RSS reader program using elasticsearch. We register 300-500 entries by the instance on average. Again, user write/and requests to elasticsearch is around 250-300 by the instance.
System works on 2 replicas and 5 shards. We have a lot of performance issues. Our CPU usage for both elastic's results is on 65-70% mark.
We communicate to elastic via C# (nest).
Has anyone ever had a similar problem improving the performance?
Our servers on aws are m1.large and m2.xlarge
How many nodes do you have in your cluster?
We register 300-500 entries by the instance on average. Again, user write/and requests to elasticsearch is around 250-300 by the instance.
I don't understand what that means. How many indices do you have? How many documents are you indexing per second (or hour, or day...)? How many search queries are being submitted per time unit?
System works on 2 replicas and 5 shards.
Why do you have 2 replicas?
© 2020. All Rights Reserved - Elasticsearch
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant logo are trademarks of the Apache Software Foundation in the United States and/or other countries.