How does Azure Data Explorer compare to Log Analytics and Azure Data Factory for log analysis with KQL and Power BI integration?

NK, SatishKumar 71 Reputation points
2023-05-29T13:11:58.5+00:00

"I'm currently using Log Analytics and Azure Data Factory to capture, transform, and analyze logs for my application on Azure. However, this solution is becoming too costly for my budget. I'm considering using Azure Data Explorer as a single solution for my log analysis needs. I'm currently using ADF just for lookup and adding additional attributes, and then storing the transformed logs back to Log Analytics Workspace as a custom table. Finally, I'm running KQL queries to power BI reports. Has anyone else used ADX for this purpose? How does it compare to Log Analytics and ADF in terms of cost, scalability, and ease of use? Are there any limitations or challenges that I should be aware of before making the switch?"

Azure Monitor
Azure Monitor
An Azure service that is used to collect, analyze, and act on telemetry data from Azure and on-premises environments.
3,014 questions
Azure Data Explorer
Azure Data Explorer
An Azure data analytics service for real-time analysis on large volumes of data streaming from sources including applications, websites, and internet of things devices.
507 questions
Azure Data Factory
Azure Data Factory
An Azure service for ingesting, preparing, and transforming data at scale.
10,134 questions
{count} votes

1 answer

Sort by: Most helpful
  1. David Broggy 5,701 Reputation points MVP
    2023-05-29T13:59:20.29+00:00

    Hi Satish,

    Once you're ok with the cost of the ADX instance the cost is quite reasonable, since you can use blob storage for your data.

    So I'd suggest calculating your monthly cost for an adx instance (eg. .095x24hoursx30days=$68US?)

    And then add on top of that your storage needs and you'll have a close estimate.

    If speed isn't critical (and it shouldn't be for scheduled reporting) consider using blob storage since you can still query your data using kql.

    If your reports contain a very large amount of data you'll want to do some performance testing to ensure the query times aren't an issue.

    reference:

    https://azure.microsoft.com/en-us/pricing/details/data-explorer/

    1 person found this answer helpful.