Performance

If an integration needs to make lots of performance-sensitive requests to our APIs in a short timespan, then it should always use the optimal caching strategy as described in this article.

Cache Header

All responses from our APIs include header x-server-cache-id which contains an ID for the node that processed your request.
If you temporarily store and mirror the last cache id received back to us on all consecutive requests, then you will be guaranteed the maximum cache hit rate in our end, providing you with the best possible performance.
The header name should be the same on your requests, x-server-cache-id

SQL stats

Every response from our APIs includes two headers:

HeaderDescription
npr-sql-statements-executed-apiThe number of SQL statements executed by the API
npr-sql-rows-read-apiThe number of rows read by the API

They provide valuable insights into the behavior of our backend if troubleshooting performance, and for endpoints that utilize caching you should notice a significant improvement.