Join us for an expert-led overview of the tools and concepts you'll need to pass exam PL-300. The first session starts on June 11th. See you there!
Get registeredJoin us at FabCon Vienna from September 15-18, 2025, for the ultimate Fabric, Power BI, SQL, and AI community-led learning event. Save €200 with code FABCOMM. Get registered
I am looking for benchmark statistics on the volume, range of data, and performance capabilities of KQL for handling high-velocity data. Specifically, I would like to know the numeric values related to the data volume KQL can process efficiently, including throughput, latency, and any scalability metrics.
Solved! Go to Solution.
Latency, ahahaha. Expect a minimum latency of 8 minutes before the inserted data is retrievable.
Hi @Nalapriya ,
Thanks for lbendlin's reply!
And @Nalapriya , There may not be any document that specifically describes the values of these attributes.
Generally speaking:
Data Ingestion Rates: Azure Data Explorer can ingest data at rates of up to 200 MB/sec per node. In large clusters, this can scale linearly, allowing ingestion rates in the range of GBs per second.
Data Storage: ADX can handle petabytes of data efficiently.
Latency: Query latency is typically in the range of milliseconds to seconds, depending on the complexity of the query and the volume of data being scanned. Simple queries on indexed data can return results in milliseconds.
Horizontal Scaling: ADX clusters can scale out by adding more nodes, allowing for linear scaling of both ingestion and query performance. A single cluster can scale to hundreds of nodes.
I have found several relevant official documents for you, hoping that they will be helpful to you:
Query limits - Kusto | Microsoft Learn
Eventhouse and KQL Database consumption - Microsoft Fabric | Microsoft Learn
Here are some commands to set these properties:
Request limits policy - Kusto | Microsoft Learn
Also these ones:
I also found a blog that has some guidance on how to do your own KQL-related performance testing, which I hope will be helpful to you:
Evaluate query performance of Azure Data Explorer | Microsoft Community Hub
Best Regards,
Dino Tao
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Hi @Nalapriya ,
Thanks for lbendlin's reply!
And @Nalapriya , There may not be any document that specifically describes the values of these attributes.
Generally speaking:
Data Ingestion Rates: Azure Data Explorer can ingest data at rates of up to 200 MB/sec per node. In large clusters, this can scale linearly, allowing ingestion rates in the range of GBs per second.
Data Storage: ADX can handle petabytes of data efficiently.
Latency: Query latency is typically in the range of milliseconds to seconds, depending on the complexity of the query and the volume of data being scanned. Simple queries on indexed data can return results in milliseconds.
Horizontal Scaling: ADX clusters can scale out by adding more nodes, allowing for linear scaling of both ingestion and query performance. A single cluster can scale to hundreds of nodes.
I have found several relevant official documents for you, hoping that they will be helpful to you:
Query limits - Kusto | Microsoft Learn
Eventhouse and KQL Database consumption - Microsoft Fabric | Microsoft Learn
Here are some commands to set these properties:
Request limits policy - Kusto | Microsoft Learn
Also these ones:
I also found a blog that has some guidance on how to do your own KQL-related performance testing, which I hope will be helpful to you:
Evaluate query performance of Azure Data Explorer | Microsoft Community Hub
Best Regards,
Dino Tao
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.
Latency, ahahaha. Expect a minimum latency of 8 minutes before the inserted data is retrievable.