unified data analytics
Overall Satisfaction with Databricks Data Intelligence Platform
Databricks is used to run all the data engineering, data science, analytics jobs related to spark , feature engineering jobs of data science and all the model training jobs of datascience. Databricks SQLwarehouse is also used as the main compute for serving the analytics powering the queries between 10 minutes to 60 minutes
Pros
- Databricks provides an amazing notebook which serves and the backbone of unified analytics for all things data engineering, data science, and analytics
- the performance of Databricks SQL Warehouse is top notch and very hard to beat by many competitors , and amazing performance optimization for delta file format
- Databricks' support for iceberg format is also very good allowing for no vendor lock - in for delta format
Cons
- The Unity catalog is restrictive in terms of policy definition due to user-defined functions compared to market alternatives like Apache Ranger or Open Policy Agent.
- Unity Catalog has a table quota limitation where a schema with a higher number of tables requires increasing the table quota repeatedly , which is not an issue if you use Hive Metastore.
- SQL Warehouse as a solution is amazing and performant, but there should have been some support to add Spark plugins or extensions so that there is some room for changes or adding custom dependencies
- the SLA has reduced due to the performance improvement when we migrated from oss spark to Databricks.
- the notebook is very well polished with lots of features , boosting our analyst productivity
- The Databricks workflows are a powerful tool to create complex dependency graphs and dags in one place
Databricks notebook give a good managed solution to all of these solutions combined with minimal maintenance
Do you think Databricks Data Intelligence Platform delivers good value for the price?
Yes
Are you happy with Databricks Data Intelligence Platform's feature set?
Yes
Did Databricks Data Intelligence Platform live up to sales and marketing promises?
Yes
Did implementation of Databricks Data Intelligence Platform go as expected?
Yes
Would you buy Databricks Data Intelligence Platform again?
Yes

Comments
Please log in to join the conversation