What is Timestream's "Query" rate limit when "Query Insights" IS NOT enabled?

Question

Hi AWS Community,

I have a few questions regarding Timestream’s Query rate limits and scaling behavior:

1. The documentation states that when Query Insights is enabled, the query rate limit is 1 TPS ([AWS Docs](https://docs.aws.amazon.com/timestream/latest/developerguide/ts-limits.html#system-limits)). However, what is the query rate limit when Query Insights is NOT enabled?
1. My service is experiencing query throttling, and I would like to understand how Timestream scales under increased traffic. Specifically:
    1. Does Timestream "query" automatically scale to accommodate higher query loads?
    1. If so, how long does it take for Timestream to scale to support a higher query rate?
1. Is there a way to increase the query rate limit proactively before the actual request rate increases? This would help us avoid throttling issues that impact our customers.

Here is an example of the error we are encountering:
```
(ThrottlingException) when calling the Query operation (reached max retries: 4): Request rate limit exceeded
```

Thank you for your help!

re:Post Agent · Answer

The documentation does not explicitly state a specific query rate limit for Amazon Timestream when Query Insights is not enabled. The 1 query per second (QPS) limit is specifically mentioned for when Query Insights is enabled.

Regarding scaling and throttling:

1. Timestream does have a mechanism for scaling query capacity, but it's not an automatic process tied directly to query rate. Instead, Timestream uses Timestream Compute Units (TCUs) to measure and allocate compute capacity for queries. The service allocates TCUs on-demand based on query complexity and the amount of data being processed.

2. The scaling of query capacity in Timestream is not instantaneous. It's controlled by the MaxQueryTCU setting, which acts as a ceiling for how much the compute capacity can scale. This setting can be adjusted from 4 to 1000 TCUs.

3. Yes, you can proactively increase the query capacity to help avoid throttling issues. This is done by increasing the MaxQueryTCU setting. It's recommended to set the MaxQueryTCU at least 4-16 TCUs higher than your peak consumption. You should consider increasing MaxQueryTCU if:
   - Your peak query consumption is approaching or reaching the current configured maximum.
   - Your queries are returning a 4xx error with the message "MaxQueryTCU exceeded".
   - You anticipate a planned increase in your workload.

The throttling error you're encountering suggests that you may need to increase your MaxQueryTCU setting. Monitor your QueryTCU CloudWatch metric to determine the required compute units and adjust the MaxQueryTCU accordingly.

It's important to note that even with these adjustments, there can still be some delay in scaling, and you might experience brief periods of throttling during sudden traffic spikes. Timestream's scaling is designed to balance performance with cost management, so it doesn't instantly scale to match every traffic fluctuation.  
**Sources**  
[Timestream Compute Unit (TCU) - Amazon Timestream](https://docs.aws.amazon.com/timestream/latest/developerguide/tcu.html)  
[QueryInsights - Amazon Timestream](https://docs.aws.amazon.com/timestream/latest/developerguide/API_query_QueryInsights.html)  
[Query - Amazon Timestream](https://docs.aws.amazon.com/timestream/latest/developerguide/API_query_Query.html)

What is Timestream's "Query" rate limit when "Query Insights" IS NOT enabled?

Relevant content