Azure Databricks vs. pandas

Suppose you have multiple data sources and you want to bring the data into one place, transform it and make it into a data model. Azure Databricks is a perfectly suited solution for this. Leverage spark JDBC or any external cloud based tool (ADG, AWS Glue) to bring the data into a cloud storage. From there, Azure Databricks can handle everything. The data can be ingested by Azure Databricks into a 3 Layer architecture based on the delta lake tables. The first layer, raw layer, has the raw as is data from source. The enrich layer, acts as the cleaning and filtering layer to clean the data at an individual table level. The gold layer, is the final layer responsible for a data model. This acts as the serving layer for BI For BI needs, if you need simple dashboards, you can leverage Azure Databricks BI to create them with a simple click! For complex dashboards, just like any sql db, you can hook it with a simple JDBC string to any external BI tool.

Incentivized

Pandas are great for quick and relatively simple analytics and visualizations

Pandas work well for exploratory ad-hoc analytic work

But , We had little success in implementing complicated predictive analytics. And large data sizes can be a problem.

Incentivized

• SQL
• Data management
• Data access

Incentivized

• It is easy to do statistical analysis
• It is easy to clean the data
• It is easy to produce graphs and charts to visualize

Incentivized

• Their pipeline workflow orchestration is pretty primitive. Lacks some common features
• Workspace UI and navigation requires steep learning curve
• Personally, I am not fond of their autosave feature. Its dangerous for production level notebooks scripts

Incentivized

• There are a lot of libraries and ways to do visualization. Sometimes it is very confusing.
• Error handling can be a challenge. Sometimes the error messages do not provide valuable clues for the debugging.
• In our case, there are a bunch of different frameworks and libraries working together. I would rather work with one framework, well tuned for my use case

Incentivized

Based on my extensive use of Azure Databricks for the past 3.5 years, it has evolved into a beautiful amalgamation of all the data domains and needs. From a data analyst, to a data engineer, to a data scientist, it jas got them all! Being language agnostic and focused on easy to use UI based control, it is a dream to use for every Data related personnel across all experience levels!

Incentivized

Over the years, we tried a lot of different frameworks and tools, homegrown and commercial. Pandas provide the best results.

It is lightweight, flexible and easy to implement.

Incentivized

Against all the tools I have used, Azure Databricks is by far the most superior of them all! Why, you ask? The UI is modern, the features are never ending and they keep adding new features. And to quote Apple, "It just works!" Far ahead of the competition, the delta lakehouse platform also fares better than it counterparts of Iceberg implementation or a loosely bound Delta Lake implementation of Synapse

Incentivized

All these frameworks are great for gathering data and providing some initial analysis. But for real performance debugging work one needs more than tools provided by this tools. That's where the pandas excel.

Incentivized

• Helped reduce time for collecting data
• Reduced cost in maintaining multiple data sources
• Access for multiple users and management of users/data in a single platform

Incentivized

• Performance debugging was time consuming and mostly poorly automated exploratory process. Once we started use pandas for these tasks, it really moved the needle. Pandas are instrumental to provide actionable insights. As a result we were able to improve notably cloud software resource utilization and performance
• Analytics implemented with pandas allow us to detect and. address problems in our APIs before they are notable to our customers

Incentivized