0

We have integrated our application with AWS API call for different services. However search/describe on some of the services have high response times (15-20 seconds) like security groups, vpc and subnet search.

ami search/describe takes even more time (>20 S) even for single AMI search.

is it the standard behavior? to meet our guidelines, we are caching some of the data but I am not sure if there is any other solution.

[UPDATE]

Thank you everyone for your suggestions. After investigating little deeper we found couple of issues. Our proxy servers are causing network latency with all external calls. Since AWS describe calls contain more data, we are observing the clear difference in response times compared to all other API interactions. Other problem is wrongly formatted Filter queries. We fixed both of them and able to see good performance improvement. Thank you for your insights.

ac184
  • 811
  • 2
  • 10
  • 19
  • 1
    Sounds highly unusual to me. I never see that level of latency for API calls. Maybe if you share some code someone can help see the problem. – dmulter Jun 28 '18 at 17:00
  • 1
    Are any of your calls to AWS API coming back in under 15 seconds? – Matt D Jun 28 '18 at 22:24

1 Answers1

1

AMI searches are impacted by publicly-available images. Make sure you specify an Owner or it will search thousands of images that aren't yours.

API call duration is often proportional to the number of resources you have.

Try making calls to regions that have no resources. That should return within milliseconds. If it takes a long time, then something is strange with your network connection.

John Rotenstein
  • 241,921
  • 22
  • 380
  • 470