At max capacity, which pricing model tcu-based or bytes-scanned allow a higher concurrent query limit?

Based on the information available, the TCU-based pricing model allows for a higher concurrent query limit compared to the bytes-scanned model. With the TCU-based model, users can configure a maximum number of Timestream Compute Units (TCUs) for their account, which directly impacts the number of concurrent queries that can be processed.

The TCU-based model allows users to set a MaxQueryTCU limit, which can be increased to support more concurrent queries. For example, in a simulated infrastructure monitoring use case, 8 TCUs were able to support 159 queries per second (QPS) with a p99 latency of less than 200 milliseconds. This suggests that by increasing the MaxQueryTCU, users can potentially achieve even higher concurrent query limits.

In contrast, the bytes-scanned model does not provide the same level of control over compute resources for query processing. The TCU-based model offers more flexibility in scaling compute resources to handle higher concurrency, making it more suitable for workloads requiring a large number of simultaneous queries.

It's important to note that while the TCU-based model allows for higher concurrency, users should carefully consider their specific workload requirements and cost implications when choosing between the two pricing models.
Sources
Understanding and optimizing Amazon Timestream Compute Units for efficient time series data management | AWS Database Blog