Elastic Search pricing: Pay as you go or not?

Question

Hi everyone,  
  
I would like to connect an Elastic Search cluster to my graphQL API (AppSync) to be able to perform a geosearch, but I'm quite confused by the pricing model...  
  
On the AWS Elastic Search page, I find "Amazon Elasticsearch Service lets you pay only for what you use – there are no upfront costs or usage requirements." and on its pricing page, the pricing table's header says "on-demand pricing". For a t2.small in Frankfurt, it costs $0.048/h. But is this for a continuously running server (hence $30/month) or only for the time you need the server to be running (genuinely on-demand)?  
  
BR,  
Arne

Answer

Hello Arne,  
  
In respect to Amazon Elasticsearch service pricing you are charged for the hours the instance is running, data transfer costs and Amazon Elastic Block Storage (which is optional and depends on your use case for the service).  
  
In the Frankfurt region you are correct, the cost to run a t2.small.elasticsearch instance is $0.042 per hour for a continuously running instance. Please note that this is the for pricing for the instance alone, more information can be found at \[1]. Currently it is not possible to pause the instances running on Amazon Elasticsearch however, rest assured our Elasticsearch service team is working hard to introduce as many features as possible to ensure customers have the best possible experience with the service.  
More information in regards Amazon Elasticsearch pricing you can refer to the pricing calculator \[2], which allows you to gain estimates of monthly costs with your chosen configuration for Elasticsearch.   
  
I hope I was able to answer your question!  
  
Best Regards,  
Shane-AWS

References:  
\[1] Elasticsearch Service Pricing Model - https://aws.amazon.com/elasticsearch-service/pricing/  
\[2] AWS Pricing Calculator - https://calculator.aws/#/

Elastic Search pricing: Pay as you go or not?

Relevant content